Multi-frequency intelligent bioelectric current cooperative modulation method

Abstract

The invention belongs to the technical field of biomedical engineering, and discloses a multi-frequency intelligent bioelectric current cooperative modulation method. The method comprises the following steps: collecting human physiological state data of each sub-treatment area in real time, performing bioelectrical impedance analysis and connectivity analysis respectively, and constructing a differentiated bioelectrical impedance area; comprehensively evaluating the tissue penetration degree of each bioelectrical impedance region; collecting treatment characteristic data, combining the treatment characteristic data with the tissue penetration degree, extracting the energy conduction efficiency of each bioelectrical impedance region, and generating an energy conduction efficiency curve; performing quantitative evaluation on the tissue response factors of each bioelectrical impedance region, and generating a response type value matrix; integrating the energy conduction efficiency curve, the response type value matrix and the treatment characteristic data, and dynamically formulating a physiotherapy parameter optimal configuration scheme; the adaptive capacity and the application effect of the bioelectric current in refined regulation and personalized treatment can be improved, so that the treatment effect and the comfort are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of biomedical engineering technology, and more specifically, to a multi-frequency intelligent biocurrent collaborative modulation method. Background Art

[0002] With the widespread application of electrical stimulation technology in medical rehabilitation, beauty care, sports rehabilitation and other fields, physical therapy equipment based on biocurrent has gradually become the mainstream non-drug intervention method; especially in terms of neural regulation, muscle function recovery and microcirculation improvement, electrical stimulation technology has shown good clinical prospects due to its non-invasiveness, controllability and good biocompatibility; traditional bioelectric stimulation methods mostly use fixed frequency and single current output methods, which often ignore individual differences in the human body, changes in physiological state and the dynamic characteristics of tissue response to current, resulting in unstable electrical stimulation effects, poor targeting and poor comfort, and it is difficult to achieve the synergistic effect of multi-frequency and zoned treatment in the same treatment process, which limits the depth and coverage of treatment; therefore, in order to improve the efficiency of physical therapy and individual adaptability, an intelligent biocurrent modulation method is urgently needed to achieve more accurate, efficient and intelligent treatment plans.

[0003] The patent with publication number CN111773541A discloses a pair of smart shaping pants based on biocurrent muscle rehabilitation training and an implementation method; it includes: a number of electrode sheets are fixed on the inner side of the smart shaping pants, and the electrode sheets correspond to different muscle groups of the human body. Each electrode sheet is connected to the corresponding voltage generation module and drive module in the controller through a corresponding wire. Several voltage generation modules and several drive modules are connected to the control module. The control module is connected to the drive module through the voltage generation module. The drive module generates biocurrent on the electrode sheet through the wire and acts on the human body; after turning on the power, different voltage generation modules and drive modules are selected through the control module, and different pulse wave signals are adjusted so that the electrode sheets at different muscle groups generate different biocurrent signals.

[0004] However, although the above-mentioned technology can realize the intelligent modulation of biocurrent, it mainly relies on the combination of preset electrode sheets and fixed drive modules for muscle stimulation. It does not take into account the individual differences and dynamically changing physiological states of different users, and lacks the real-time perception and intelligent adaptation capabilities of the user's physiological characteristics; and the regulation method of the above-mentioned technology is mainly based on basic pulse waveform control, which does not reflect the differentiated current modulation capabilities for different regions; therefore, the above-mentioned technology has deficiencies in both refined control and personalized treatment, which makes it difficult to achieve precise biocurrent control for different users, different treatment stages or different physiological states in actual applications, thereby affecting the individualization, precision and long-term stability of the treatment effect.

[0005] In view of this, the present invention proposes a multi-frequency intelligent biocurrent coordinated modulation method to solve the above problems. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned objectives, the present invention provides the following technical solutions: a multi-frequency intelligent biocurrent collaborative modulation method, comprising: Step S1: Real-time collection of human physiological status data of each sub-treatment area within the treatment area, where the human physiological status data includes skin physiological data, tissue structure data, muscle physiological data, and blood circulation data; Step S2: Fusing muscle physiological data with blood circulation data, performing bioelectrical impedance analysis on each sub-treatment area, performing connectivity analysis on all sub-treatment areas based on the bioelectrical impedance analysis results, and constructing differentiated bioelectrical impedance areas; Step S3: Based on the skin physiological data and tissue structure data, the sensitivity of human tissue in each bioelectrical impedance area is deeply characterized, and combined with the blood circulation data, the tissue penetration degree of each bioelectrical impedance area is comprehensively evaluated; Step S4: collecting treatment characteristic data in real time, and combining it with the degree of tissue penetration, extracting the energy conduction efficiency of each bioelectrical impedance area, and generating an energy conduction efficiency curve; Step S5: combining skin physiological data and blood circulation data to quantitatively evaluate the tissue response factors of each bioelectrical impedance area, and classifying and labeling the tissue response factors based on a threshold decision mechanism to generate a response type value matrix; Step S6: Integrate the energy conduction efficiency curve, the response type value matrix, and the treatment characteristic data to dynamically formulate an optimal configuration plan for the therapy parameters, and coordinately modulate the biocurrents acting on different bioelectrical impedance areas based on the optimal configuration plan for the therapy parameters.

[0007] Furthermore, the skin physiological data includes skin surface temperature, skin moisture content, skin pH value, sebum secretion and skin elasticity coefficient; the tissue structure data is subcutaneous fat thickness; the muscle physiological data is muscle tension; the blood circulation data includes blood flow rate and blood oxygen saturation; The method of performing bioelectrical impedance analysis on each sub-treatment area includes: The same muscle tension and blood flow rate in the corresponding sub-treatment area are used as a set of impedance analysis data, and the impedance analysis data corresponds to the sub-treatment area one by one; each set of impedance analysis data is input into the trained impedance prediction model to predict the electrical impedance value corresponding to each sub-treatment area.

[0008] Furthermore, the method for constructing differentiated bioelectrical impedance regions includes: A clustering algorithm is used to cluster the electrical impedance values of all sub-treatment areas to obtain two impedance categories; the electrical impedance values in the two impedance categories are averaged to obtain the electrical impedance mean; the two electrical impedance means are compared, and the impedance category with the larger electrical impedance mean is marked as the high-impedance category, and the impedance category with the smaller electrical impedance mean is marked as the low-impedance category; if the electrical impedance value is in the high-impedance category, the area category of the corresponding sub-treatment area is defined as the high-impedance area; if the electrical impedance value is in the low-impedance category, the area category of the corresponding sub-treatment area is defined as the low-impedance area; The adjacent areas corresponding to each sub-treatment area are taken as a set of adjacent sets, and the adjacent sets correspond to the sub-treatment areas one by one. In each set of adjacent sets, the adjacent areas with the same area category as the corresponding sub-treatment area are screened out and marked as connected areas. An associated edge is established between each connected area and the corresponding sub-treatment area, and all sub-treatment areas are taken as nodes. An associated graph is constructed based on the nodes and associated edges. In the associated graph, the depth-first search method is used to search for connected components and obtain Connected components are formed, and the sub-treatment areas corresponding to the nodes in each connected component are merged to construct a bioelectrical impedance area. The bioelectrical impedance area corresponds to the connected components one by one. is an integer greater than 0; if the region categories of all adjacent regions in the adjacent set are different from the corresponding sub-treatment region, the corresponding sub-treatment region will be regarded as a bioelectrical impedance region.

[0009] Furthermore, the method for in-depth characterization of the sensitivity of human tissue in each bioelectrical impedance region includes: Corresponding neural network encoder models are constructed for skin surface temperature, subcutaneous fat thickness and skin moisture content respectively; based on the neural network encoder model, the skin surface temperature, subcutaneous fat thickness and skin moisture content of each bioelectrical impedance area are mapped into corresponding high-dimensional feature vectors respectively; the same high-dimensional feature vectors of the corresponding bioelectrical impedance area are regarded as a group of feature sets, and each group of feature sets is input into the pre-constructed tensor fusion network in turn for a three-modal tensor product operation to generate a sensitivity fusion vector corresponding to each bioelectrical impedance area; each sensitivity fusion vector is sequentially subjected to dimensionality reduction operation through the pre-constructed fully connected neural network to obtain the sensitivity feature vector corresponding to each bioelectrical impedance area as an in-depth representation of human tissue sensitivity.

[0010] Furthermore, the method for comprehensively evaluating the tissue penetration degree of each bioelectrical impedance region includes: A corresponding neural network encoder model is constructed for blood oxygen saturation. Based on the neural network encoder model, the blood oxygen saturation of each bioelectrical impedance area is mapped to a corresponding blood oxygen feature vector. A weight set is preset, and the weight set includes weight coefficients corresponding to the sensitivity feature vector and the blood oxygen feature vector. Based on the weight set, the sensitivity feature vector and the blood oxygen feature vector of each bioelectrical impedance area are weightedly summed to obtain the tissue penetration degree of each bioelectrical impedance area.

[0011] Furthermore, the treatment characteristic data includes area and depth information of each bioelectrical impedance region; The method for extracting the energy conduction efficiency of each bioelectrical impedance region includes: A corresponding tissue conductivity model and current density distribution model are constructed for each bioelectrical impedance region; the tissue conductivity model and current density distribution model corresponding to the bioelectrical impedance region are fused to obtain a fused simulation model; The electrical impedance value, area, depth information, and tissue penetration degree corresponding to each bioelectrical impedance region are input into the corresponding fusion simulation model, and multiple biocurrents are set as input for each fusion simulation model. Numerical simulation technology is used to simulate each fusion simulation model in turn, and the energy conduction ratio corresponding to each biocurrent is output. All energy conduction ratios with the same corresponding fusion simulation model are averaged to obtain the simulation efficiency of each bioelectrical impedance region. The area, depth information, and tissue penetration degree of each bioimpedance region are sequentially standardized to obtain standardized data for each bioimpedance region; the standardized data includes the standardized area, standard depth, and standard penetration degree; the analytical efficiency of each bioimpedance region is calculated based on the functional relationship between each type of data within the standardized data; the simulation ratio of each bioimpedance region is calculated based on the simulation efficiency, and the analytical ratio of each bioimpedance region is calculated based on the analytical efficiency; and the sum of the simulation ratio and analytical ratio for the same bioimpedance region is used as the energy conduction efficiency of the corresponding bioimpedance region; The method for generating an energy conduction efficiency curve comprises: The energy conduction efficiency of each bioelectrical impedance area is sorted in ascending order to generate an efficiency sequence; according to the positive order of the efficiency sequence, the bioelectrical impedance area corresponding to each energy conduction efficiency is set with a numerical label in ascending order and marked as a regional label; the regional label corresponding to each energy conduction efficiency is used as the horizontal axis index, and the corresponding energy conduction efficiency is used as the vertical axis value to generate an energy conduction efficiency curve.

[0012] Furthermore, the step of quantitatively evaluating the tissue response factor of each bioelectrical impedance area includes: Step S501: The skin pH value, sebum secretion, skin elasticity coefficient and blood flow rate of each bioelectrical impedance area are taken as a set of impact analysis data; Step S502: construct multiple fuzzy sets for each data in the impact analysis data; Step S503: converting each set of impact analysis data into the membership degree of each corresponding fuzzy set through fuzzification technology; Step S504: defining fuzzy rules; Step S505: Match each set of fuzzified impact analysis data with fuzzy rules respectively, and perform fuzzy reasoning using a fuzzy reasoning method to obtain the fuzzy reasoning results corresponding to each bioelectrical impedance area. The fuzzy reasoning results are the membership of each response level, which includes high level, medium level, and low level. Step S506: Set the factor interval. The range of the factor interval is ; Divide the factor interval evenly into three level intervals, and the level intervals correspond one-to-one to the levels in the response level; take the mean of the maximum and minimum values of each level interval as the interval mean of the corresponding level interval; Step S507: Based on the interval mean, perform weighted summation on the membership of each response level corresponding to each bioelectrical impedance area to obtain the total response value of each bioelectrical impedance area; add the membership of each response level corresponding to each bioelectrical impedance area in sequence to obtain the total membership; and use the ratio of the total response value of each bioelectrical impedance area to the corresponding total membership as the tissue response factor of each bioelectrical impedance area.

[0013] Furthermore, the method for generating a response type value matrix includes: According to the area type corresponding to the treatment area, the corresponding response threshold is obtained from a pre-constructed threshold set; the threshold set includes response thresholds corresponding to different area types, and each tissue response factor is compared with the response threshold respectively; the tissue response factor with a value greater than or equal to the response threshold is marked as a positive response factor, and the tissue response factor with a value less than the response threshold is marked as a negative response factor; different digital labels are set for the positive response factor and the negative response factor, and marked as response type values; according to the tissue response factor of each bioelectrical impedance area, the corresponding response type value is obtained, and a response type value matrix is generated based on the response type value.

[0014] Furthermore, the step of dynamically formulating the optimal configuration plan for physical therapy parameters includes: Step S601: Acquire parameter ranges based on the energy conduction efficiency curve, the response type value matrix, the treatment characteristic data, and the electrical impedance values of each bioelectrical impedance region; Step S602: Construct based on parameter range Group parameter collection, is an integer greater than 1; Step S603: From Randomly select from the group parameter set Group parameter set and All group parameter sets are treated as particles to construct a particle group, and the number of iterations is set to 0; Step S604: determining the best particle and the worst particle in the particle group; Step S605: Calculate the charge value corresponding to each particle in the particle group; Step S606: Calculating the guiding electromagnetic force of each particle in the particle group based on the charge value; Step S607: updating each particle in the particle group according to the guiding electromagnetic force; Step S608: locally updating the optimal particle in the particle group; Step S609: Compare the number of iterations with a preset iteration threshold. If the number of iterations is less than the iteration threshold, increment the number of iterations and return to step S604. If the number of iterations is greater than or equal to the iteration threshold, proceed to step S610. Step S610: obtaining the optimal particle in the particle group, marking the parameter set corresponding to the optimal particle as the optimal set, and using the optimal set as the optimal configuration scheme for the physical therapy parameters.

[0015] Furthermore, in step S601, the parameter range includes the value range corresponding to each parameter in the physical therapy parameters; The method for obtaining the parameter range is as follows: the energy conduction efficiency curve, the response type value matrix, the treatment characteristic data, and the electrical impedance values of each bioelectrical impedance region are used as range analysis data, and the range analysis data are input into the trained range prediction model to predict the corresponding parameter range; In step S602, construct The method of grouping parameter sets is: randomly select a value from each value range within the parameter range and construct a set of parameter sets; and so on. Group parameter collection, The set of group parameters are all different; In step S604, the method for determining the best particle and the worst particle in the particle group includes: The range analysis data and the parameter set corresponding to each particle are regarded as a set of effect analysis data, and the effect analysis data and the parameter set correspond one-to-one; each set of effect analysis data is input into the trained effect prediction model to predict the physical therapy effect corresponding to each set of effect analysis data; the physical therapy effect corresponding to each particle is compared respectively, and the particle with the largest physical therapy effect is marked as the optimal particle, and the particle with the smallest physical therapy effect is marked as the worst particle.

[0016] The technical effects and advantages of the multi-frequency intelligent biocurrent coordinated modulation method of the present invention are as follows: By collecting multi-dimensional human physiological state data in real time, differentiated bioelectrical impedance areas are constructed, and the tissue characteristics and electrophysiological environment of the target treatment area are accurately portrayed; based on in-depth analysis of human tissue sensitivity and tissue penetration, the response characteristics of each bioelectrical impedance area to bioelectric current are dynamically evaluated, and energy conduction efficiency curves and response type value matrices are generated, providing a reliable basis for subsequent personalized regulation; an intelligent optimization algorithm is used in combination with treatment characteristic data to dynamically generate the optimal configuration scheme of physical therapy parameters, which can modulate bioelectric currents of different frequencies, intensities and other parameters for different bioelectrical impedance areas, and realize precise and coordinated multi-frequency treatment; it realizes real-time perception and intelligent response to changes in physiological state during treatment, effectively solving the problems of poor targeting and rough regulation in traditional bioelectrical current treatment, and significantly improving the adaptability and application effect of bioelectric current in refined regulation and personalized treatment, thereby effectively improving the treatment effect and comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a multi-frequency intelligent biocurrent collaborative modulation method according to embodiment 1 of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] Example 1 See also Figure 1 As shown, this embodiment provides a multi-frequency intelligent biocurrent coordinated modulation method, the method comprising: Step S1: Real-time collection of human physiological status data of each sub-treatment area within the treatment area, where the human physiological status data includes skin physiological data, tissue structure data, muscle physiological data, and blood circulation data.

[0020] The treatment area is the overall range of the target body part for bioelectric current treatment; for example, if it is used for rehabilitation training, the treatment area is such as the thigh, waist, etc.; if it is used for medical physiotherapy, the treatment area is such as the shoulders, neck, waist and back, etc.; the sub-treatment area is a smaller area obtained by further subdividing the treatment area according to the physiological structure or treatment needs; for example, the waist can be subdivided into sub-treatment areas such as the left and right erector spinae muscles, small muscle groups, and lumbar vertebrae; the thigh can be subdivided into sub-treatment areas such as the quadriceps femoris, biceps femoris, and adductor muscles; the treatment area and sub-treatment area are set and adjusted by doctors, rehabilitation therapists or other professionals according to the user's individual situation, rehabilitation goals and specific physiological characteristics.

[0021] Skin physiological data includes skin surface temperature, skin moisture content, skin pH value, sebum secretion and skin elasticity coefficient; tissue structure data includes subcutaneous fat thickness; muscle physiological data includes muscle tension; blood circulation data includes blood flow rate and blood oxygen saturation; Among them, because the values of skin surface temperature, skin moisture content, subcutaneous fat thickness, muscle tension, and blood flow velocity vary greatly in different parts of the human body, reflecting the local physiological state, distributed collection is performed, that is, each sub-treatment area is collected. Skin pH value, sebum secretion, skin elasticity coefficient, and blood oxygen saturation are relatively uniform over a large range and are insensitive to local differences. Therefore, holistic collection is performed, that is, only the entire treatment area is collected. Skin surface temperature is the heat of the skin surface, reflecting the local blood circulation and metabolic state, and is measured and obtained in each sub-treatment area using an infrared thermometer or temperature sensor; skin moisture content is the water content of the skin surface, reflecting the moisture and barrier function of the skin, and is measured and obtained in each sub-treatment area using a skin moisture tester; skin pH is the acidity and alkalinity of the skin surface, reflecting the skin state, and is measured and obtained in the treatment area using a skin pH tester; sebum secretion is the amount of oil secreted by the sebaceous glands, which affects the skin barrier protection, and is measured and obtained in the treatment area using a sebum meter or oil paper adsorption method; skin elasticity is the skin's ability to recover when subjected to mechanical stretch or pressure, reflecting the skin's flexibility, and is measured and obtained using a skin elasticity tester The treatment area is measured and obtained; the subcutaneous fat thickness is the thickness of the fat tissue under the skin, which reflects the local fat distribution and body shape characteristics, and is measured and obtained for each sub-treatment area using a skinfold thickness meter or an ultrasonic thickness gauge; muscle tension is the tightness of the muscle contraction state, which reflects the muscle activity level and fatigue state, and is measured and obtained for each sub-treatment area using a muscle hardness meter or an electromyography device; the blood flow rate is the flow speed of blood in the capillaries and microcirculation, which reflects the local circulation state, and is measured and obtained for each sub-treatment area using a laser Doppler blood flowmeter or an ultrasonic Doppler blood flowmeter; the blood oxygen saturation is the ratio of oxygen to hemoglobin in the blood, which reflects the tissue oxygenation state, and is measured and obtained from the user using a pulse oximeter.

[0022] Step S2: Fusing muscle physiological data and blood circulation data, performing bioelectrical impedance analysis on each sub-treatment area, performing connectivity analysis on all sub-treatment areas based on the bioelectrical impedance analysis results, and constructing differentiated bioelectrical impedance areas.

[0023] Methods for performing bioelectrical impedance analysis on each sub-treatment area include: The muscle tension and blood flow rate of the corresponding sub-treatment area are used as a set of impedance analysis data, and the impedance analysis data corresponds to the sub-treatment area one by one. Each set of impedance analysis data is input into the trained impedance prediction model to predict the electrical impedance value corresponding to each sub-treatment area. The impedance prediction model is a deep neural network model. The training process of the impedance prediction model includes: Pre-collection Set different impedance analysis data, and set the corresponding electrical impedance value for each set of impedance analysis data. is an integer greater than 1; the impedance analysis data and the corresponding electrical impedance value are converted into a corresponding set of feature vectors; the electrical impedance value corresponding to the impedance analysis data is collected by those skilled in the art in the process of constructing differentiated bioelectrical impedance regions in history. Different groups of impedance analysis data are analyzed in turn using bioelectrical impedance analysis methods (such as multi-frequency bioelectrical impedance analysis, dual-frequency bioelectrical impedance analysis, AC impedance analysis, etc.) to obtain the corresponding electrical impedance value; Set the corresponding electrical impedance values for different groups of impedance analysis data in turn; Each set of eigenvectors is used as the input of the impedance prediction model. The impedance prediction model takes a set of predicted impedance values corresponding to each set of impedance analysis data as the output, and the actual impedance value corresponding to each set of impedance analysis data as the prediction target. The actual impedance value is the pre-set impedance value corresponding to the impedance analysis data. The training goal is to minimize the sum of the prediction errors of all impedance analysis data. The calculation formula of the prediction error is: ,in is the prediction error, is the group number of the eigenvector corresponding to the impedance analysis data, For the The predicted electrical impedance value corresponding to the group impedance analysis data, For the The actual electrical impedance value corresponding to the impedance analysis data is obtained; the impedance prediction model is trained until the sum of the prediction errors reaches convergence and the training is stopped.

[0024] It should be noted that the reason for integrating muscle tension and blood flow rate for bioelectrical impedance analysis is that when muscle tension increases, tissue density increases and muscle fibers are arranged more densely, so the current path is blocked, resulting in an increase in electrical impedance value, and vice versa; when blood flow rate increases, the blood content of local tissues increases, and blood, as a good conductor, enhances tissue conductivity, so the electrical impedance value decreases, and vice versa.

[0025] Methods for constructing differentiated bioelectrical impedance regions include: A clustering algorithm (such as K-means clustering, hierarchical clustering, Gaussian mixture clustering, etc.) is used to cluster the electrical impedance values of all sub-treatment areas to obtain two impedance categories; the electrical impedance values in the two impedance categories are respectively averaged to obtain the electrical impedance mean; the two electrical impedance means are compared, and the impedance category with the larger electrical impedance mean is marked as a high-impedance category, and the impedance category with the smaller electrical impedance mean is marked as a low-impedance category; if the electrical impedance value is in the high-impedance category, the area category of the corresponding sub-treatment area is defined as a high-impedance area; if the electrical impedance value is in the low-impedance category, the area category of the corresponding sub-treatment area is defined as a low-impedance area; The adjacent areas corresponding to each sub-treatment area are considered as a set of adjacent sets, and the adjacent sets correspond to the sub-treatment areas one by one. The adjacent areas of a sub-treatment area are the remaining sub-treatment areas that have boundary contact with the sub-treatment area. In each set of adjacent sets, the adjacent areas with the same area category as the corresponding sub-treatment area are selected and marked as connected areas. An associated edge is established between each connected area and the corresponding sub-treatment area, and all sub-treatment areas are regarded as nodes. An associated graph is constructed based on the nodes and associated edges. In the association graph, the depth-first search method is used to find the connected components and obtain Connected components are formed, and the sub-treatment areas corresponding to the nodes in each connected component are merged to construct a bioelectrical impedance area. The bioelectrical impedance area corresponds to the connected components one by one. is an integer greater than 0; wherein, any two sub-treatment areas in each connected component are connected by a path; it should be noted that the depth-first search method is a prior art, and the specific process will not be described in detail here; illustratively, there are five sub-treatment areas A, B, C, D, and E, wherein there is an associated edge between A and B, an associated edge between B and D, and an associated edge between C and E, therefore, A, B and D are one connected component, and C and E are another connected component; if the area categories of all adjacent areas in the adjacent set are different from those of the corresponding sub-treatment areas, the corresponding sub-treatment area will be regarded as a bioelectrical impedance area.

[0026] Step S3: Based on the skin physiological data and tissue structure data, the sensitivity of human tissue in each bioelectrical impedance area is deeply characterized, and combined with the blood circulation data, the tissue penetration degree of each bioelectrical impedance area is comprehensively evaluated.

[0027] Methods for in-depth characterization of human tissue sensitivity across bioelectrical impedance regions include: Corresponding neural network encoder models are constructed for skin surface temperature, subcutaneous fat thickness and skin moisture content respectively. The neural network encoder model is an encoder based on a neural network structure, which is used to map the input single data into the corresponding high-dimensional feature vector, thereby realizing effective feature extraction and expression of the original numerical signal; based on the neural network encoder model, the skin surface temperature, subcutaneous fat thickness and skin moisture content of each bioelectrical impedance area are mapped into the corresponding high-dimensional feature vector respectively; the same high-dimensional feature vectors of the corresponding bioelectrical impedance area are regarded as a group of feature sets, and each group of feature sets is sequentially input into a pre-constructed tensor fusion network for a three-modal tensor product operation to generate a sensitivity fusion vector corresponding to each bioelectrical impedance area; each sensitivity fusion vector is sequentially subjected to a dimensionality reduction operation through a pre-constructed fully connected neural network to obtain the sensitivity feature vector corresponding to each bioelectrical impedance area as an in-depth representation of human tissue sensitivity; It should be noted that the neural network encoder model, tensor fusion network and fully connected neural network are all existing technologies, and the specific construction process will not be elaborated here; human tissue sensitivity is the degree of tissue response to bioelectric current; skin surface temperature will affect local blood circulation and nerve ending activity, thereby affecting human tissue sensitivity; subcutaneous fat thickness acts as a natural barrier, which will affect the depth and intensity of energy conduction. The thicker the fat layer, the lower the corresponding human tissue sensitivity in the sub-treatment area, and vice versa; skin moisture content directly affects the electrical conductivity and thermal conduction properties of the tissue. Skin with high moisture content is usually more conductive and more sensitive to bioelectric current, so human tissue sensitivity is higher, and vice versa.

[0028] Methods for comprehensively evaluating the degree of tissue penetration in each bioelectrical impedance area include: A corresponding neural network encoder model is constructed for blood oxygen saturation. Based on the neural network encoder model, the blood oxygen saturation of each bioelectrical impedance region is mapped to a corresponding blood oxygen feature vector. A preset weight set is set, which includes weight coefficients corresponding to the sensitivity feature vector and the blood oxygen feature vector. The weight set is pre-set by those skilled in the art based on actual conditions. Based on the weight set, the sensitivity feature vector and the blood oxygen feature vector of each bioelectrical impedance region are weighted and summed to obtain the tissue penetration degree of each bioelectrical impedance region. The tissue penetration degree is the depth that the bioelectrical current can effectively penetrate in the bioelectrical impedance region. The higher the sensitivity of human tissue, the stronger the tissue conductivity, the more sufficient the energy coupling, and the better the penetration effect. Therefore, the tissue penetration degree is higher, and vice versa. Blood oxygen saturation reflects the blood circulation and metabolic state of local tissue. High blood oxygen saturation helps maintain a good electrophysiological environment and tissue activity, thereby improving the efficiency and depth of energy conduction in the tissue. Therefore, the tissue penetration degree is higher, and vice versa.

[0029] Step S4: Collect treatment characteristic data in real time, and extract the energy conduction efficiency of each bioelectrical impedance area in combination with the tissue penetration degree, and generate an energy conduction efficiency curve.

[0030] Treatment characteristic data includes the area and depth information of each bioelectrical impedance region; the area is the projected area of the bioelectrical impedance region on the human body surface; the depth information is the vertical depth of the bioelectrical impedance region in the human tissue, indicating the distance below the skin surface where the target tissue to be treated is located; The method to obtain the area of the region is: Obtaining the sub-region area, which is the area of the sub-treatment region, and is obtained by a person skilled in the art by directly measuring the user; counting the number of neutron treatment regions in each bioelectrical impedance region and marking it as the region number; multiplying the number of each region by the sub-region area as the region area of the corresponding bioelectrical impedance region; The depth information is obtained by those skilled in the art using electrical impedance tomography, ultrasonic imaging and other technical means.

[0031] Methods for extracting the energy conduction efficiency of each bioelectrical impedance region include: For each bioelectrical impedance region, a corresponding tissue conductivity model and current density distribution model are constructed. The tissue conductivity model is used to accurately describe the conduction characteristics of the tissue contained in the bioelectrical impedance region to bioelectrical current, and the current density distribution model is used to simulate the specific flow path and distribution state of the current in the bioelectrical impedance region. Both the tissue conductivity model and the current density distribution model are constructed by those skilled in the art based on a public tissue conductivity database, medical image information, and actual electrode layout, and in combination with electromagnetic field modeling principles and finite element numerical analysis methods. Both are prior arts, and the specific construction methods are not elaborated on here. The tissue conductivity model and current density distribution model corresponding to the same bioelectrical impedance region are fused to obtain a fused simulation model. The fused simulation model is obtained by mapping the tissue type and conductivity information in the tissue conductivity model to the spatial discrete grid within the current density distribution model. Based on the Maxwell equations under quasi-static conditions, a coupled mathematical relationship between the conductivity field and the potential field is constructed. The coupled mathematical relationship is solved using the finite element method to obtain the steady-state potential distribution, and the current density vector field is calculated based on the potential gradient to generate the fused simulation model. The electrical impedance value, regional area, depth information and tissue penetration degree corresponding to each bioelectrical impedance region are input into the corresponding fusion simulation model, and a technician in this field sets a variety of bioelectrical currents with different amplitudes, frequencies and other parameters as input for each fusion simulation model; each fusion simulation model is simulated in turn using numerical simulation technology (such as finite element analysis method, Monte Carlo simulation, etc.), and the energy conduction ratio corresponding to each bioelectrical current is output; the energy conduction ratio is the ratio of the energy passing through the bioelectrical impedance region to the total input energy under specific bioelectrical current conditions; the simulation process is to discretize or randomly sample the complex tissue structure to simulate the propagation and distribution of the bioelectrical current in the tissue; all energy conduction ratios with the same corresponding fusion simulation model are averaged to obtain the simulation efficiency of each bioelectrical impedance region; The area, depth information and tissue penetration degree of each bioimpedance region are standardized in turn (such as minimum-maximum standardization, logarithmic standardization, etc.) to obtain standardized data for each bioimpedance region; the standardized data includes standard area, standard depth and standard penetration degree; based on the functional relationship between each data in the standardized data, the analytical efficiency of each bioimpedance region is calculated; based on the region type corresponding to the treatment region (such as thigh, waist, shoulder and neck, etc.), the corresponding proportional factor is obtained from the pre-constructed proportional set; the proportional set includes proportional factors corresponding to different region types, and the value range of the proportional factor is The proportion set is pre-set by those skilled in the art according to actual conditions; the proportion factor is subtracted from 1 to obtain the reverse factor; the product of each simulation efficiency and the proportion factor is used as the simulation proportion, and the product of each analytical efficiency and the reverse factor is used as the analytical proportion; the sum of the same simulation proportion and analytical proportion in the corresponding bioelectrical impedance area is used as the energy conduction efficiency of the corresponding bioelectrical impedance area, and the energy conduction efficiency is the overall ability of the bioelectrical impedance area to conduct energy.

[0032] The expression of analytical efficiency is: ; Where, For analytical efficiency, is the influence coefficient, is the standard penetration level, For standard depth, is the standard area; the influence coefficient is pre-set by those skilled in the art according to actual conditions; the higher the tissue penetration efficiency, the more effectively the energy can penetrate the tissue, so the higher the resolution efficiency, and vice versa; since there is spatial attenuation when energy is conducted in the tissue, when the depth information increases, the energy attenuates more when penetrating deeper tissue, resulting in a decrease in resolution efficiency, and vice versa; the larger the regional area, the more dispersed the energy, and the lower the energy density per unit area, so the lower the resolution efficiency, and vice versa.

[0033] It should be understood that the purpose of extracting energy conduction efficiency by combining the fusion simulation model with mathematical expressions is to improve the overall stability and adaptability while ensuring the accuracy of the model, and to make the model more interpretable and generalizable through reasonable mathematical expressions; it not only helps to improve the prediction accuracy of the model in different bioelectrical impedance areas and different experimental conditions, but also has better adaptability and stronger universality when facing complex practical application scenarios.

[0034] Methods for generating energy transfer efficiency curves include: The energy conduction efficiency of each bioelectrical impedance area is sorted in ascending order (i.e., sorted from small to large) to generate an efficiency sequence; according to the positive order of the efficiency sequence, the bioelectrical impedance area corresponding to each energy conduction efficiency is set with successively increasing numerical labels and marked as area labels; the area label corresponding to each energy conduction efficiency is used as the horizontal axis index, and the corresponding energy conduction efficiency is used as the vertical axis value to generate an energy conduction efficiency curve.

[0035] Step S5: Combining skin physiological data and blood circulation data, quantitatively evaluate the tissue response factors of each bioelectrical impedance area, and classify and label the tissue response factors based on the threshold decision mechanism to generate a response type value matrix.

[0036] The steps for quantitatively evaluating the tissue response factor for each bioelectrical impedance area include: Step S501: The skin pH value, sebum secretion, skin elasticity coefficient and blood flow rate of each bioelectrical impedance area are taken as a set of impact analysis data, and the impact analysis data correspond to the bioelectrical impedance area one by one; Step S502: Construct multiple fuzzy sets for each data in the impact analysis data; for example, the fuzzy sets corresponding to skin pH value are acidic, neutral, alkaline, etc.; the fuzzy sets corresponding to sebum secretion are large secretion, medium secretion, small secretion, etc.; the fuzzy sets corresponding to skin elasticity coefficient are high elasticity, medium elasticity, low elasticity, etc.; and the fuzzy sets corresponding to blood flow rate are high rate, medium rate, low rate, etc.; Step S503: Each set of impact analysis data is converted into the membership of each corresponding fuzzy set using fuzzification technology. Fuzzification technology is the process of converting precise numerical values into the membership of the corresponding fuzzy set. Fuzzification technologies include triangular membership function and trapezoidal membership function. For example, if the value of sebum secretion is high, the membership of large secretion is inferred to be 0.9, the membership of medium secretion is 0.3, and the membership of small secretion is 0. Step S504: defining fuzzy rules. The fuzzy rules are defined based on expert knowledge or relevant literature. For example, if the fuzzy set is neutral, medium secretion, high elasticity, and medium rate, then the membership of inferring that the response level belongs to a high level is high; if the fuzzy set is alkaline, high secretion, low elasticity, and low rate, then the membership of inferring that the response level belongs to a low level is high. Step S505: Each group of fuzzified impact analysis data is matched with fuzzy rules respectively, and fuzzy reasoning is performed using a fuzzy reasoning method (such as the Mamdani fuzzy reasoning model, the Sugeno fuzzy reasoning model, etc.) to obtain a fuzzy reasoning result corresponding to each bioelectrical impedance area. The fuzzy reasoning result is the membership degree of each response level, which includes high level, medium level, and low level. For example, the fuzzy reasoning result has a high level membership degree of 0.6, a medium level membership degree of 0.8, and a low level membership degree of 0.1. Step S506: Set the factor interval. The range of the factor interval is , and The specific value of is set by those skilled in the art according to the actual situation; the factor interval is evenly divided into three level intervals, and the level intervals correspond to the levels in the response level one by one; the mean of the maximum and minimum values of each level interval is taken as the interval mean of the corresponding level interval; Step S507: Based on the interval mean, perform weighted summation on the membership of each response level corresponding to each bioelectrical impedance area to obtain the total response value of each bioelectrical impedance area; add the membership of each response level corresponding to each bioelectrical impedance area in sequence to obtain the total membership; take the ratio of the total response value of each bioelectrical impedance area to the corresponding total membership as the tissue response factor of each bioelectrical impedance area; the tissue response factor is a comprehensive quantitative indicator for evaluating the physiological response degree and sensitivity of human tissue when receiving bioelectrical current stimulation, reflecting the intensity of the tissue's response to external physical therapy stimulation.

[0037] Methods for generating a matrix of response type values include: According to the area type corresponding to the treatment area, the corresponding response threshold is obtained from a pre-constructed threshold set; the threshold set includes response thresholds corresponding to different area types, and the threshold set is pre-set by technical personnel in this field according to actual conditions; each tissue response factor is compared with the response threshold, and the tissue response factor with a value greater than or equal to the response threshold is marked as a positive response factor, and the tissue response factor with a value less than the response threshold is marked as a negative response factor; wherein, the positive response factor indicates that the tissue shows good acceptance and positive response to the bioelectric current, and the negative response factor indicates that the tissue shows low sensitivity, adverse reaction or potential side effect risk to the bioelectric current; different digital labels are set for the positive response factor and the negative response factor, and marked as response type values; according to the tissue response factor of each bioelectrical impedance area, the corresponding response type value is obtained, and a response type value matrix is generated based on the response type value.

[0038] Step S6: Integrate the energy conduction efficiency curve, the response type value matrix, and the treatment characteristic data to dynamically formulate an optimal configuration plan for the therapy parameters, and coordinately modulate the biocurrents acting on different bioelectrical impedance areas based on the optimal configuration plan for the therapy parameters.

[0039] The steps for dynamically developing the optimal configuration of physical therapy parameters include: Step S601: Acquire parameter ranges based on the energy conduction efficiency curve, the response type value matrix, the treatment characteristic data, and the electrical impedance values of each bioelectrical impedance region; Step S602: Construct based on parameter range Group parameter collection, is an integer greater than 1; Step S603: From Randomly select from the group parameter set Group parameter set and All group parameter sets are treated as particles to construct a particle group, and the number of iterations is set to 0; Step S604: determining the best particle and the worst particle in the particle group; Step S605: Calculate the charge value corresponding to each particle in the particle group; Step S606: Calculating the guiding electromagnetic force of each particle in the particle group based on the charge value; Step S607: updating each particle in the particle group according to the guiding electromagnetic force; Step S608: locally updating the optimal particle in the particle group; Step S609: Compare the number of iterations with a preset iteration threshold. If the number of iterations is less than the iteration threshold, increment the number of iterations and return to step S604. If the number of iterations is greater than or equal to the iteration threshold, proceed to step S610. Step S610: obtaining the optimal particle in the particle group, marking the parameter set corresponding to the optimal particle as the optimal set, and using the optimal set as the optimal configuration scheme for the physical therapy parameters.

[0040] In step S601 above, the parameter range includes the value range corresponding to each parameter of the physical therapy parameters, including but not limited to current intensity (i.e., the magnitude of the bioelectric current passing through the tissue per unit time), current waveform (i.e., the shape of the current changing over time, such as a sine wave, square wave, or triangle wave), pulse width (i.e., the duration of each electrical pulse), current frequency (i.e., the number of current pulses occurring per unit time), and treatment duration (i.e., the duration of each treatment). The method for obtaining the parameter range is: using the energy conduction efficiency curve, response type value matrix, treatment characteristic data, and the electrical impedance values of each bioelectrical impedance area as range analysis data, inputting the range analysis data into a trained range prediction model to predict the corresponding parameter range; the range prediction model is a deep neural network model, and the training process of the range prediction model is consistent with the training process of the impedance prediction model.

[0041] In the above step S602, construct The method of grouping parameter sets is: randomly select a value from each value range within the parameter range and construct a set of parameter sets; and so on. Group parameter collection, The set of group parameters are all different.

[0042] In the above step S603 , the particles in the particle group correspond to the parameter set one by one.

[0043] In step S604, the method for determining the best particle and the worst particle in the particle group includes: The range analysis data and the parameter set corresponding to each particle are regarded as a set of effect analysis data, and the effect analysis data and the parameter set correspond one-to-one; each set of effect analysis data is input into the trained effect prediction model to predict the therapeutic effect corresponding to each set of effect analysis data; wherein, the effect prediction model is a deep neural network model, and the training process of the effect prediction model is consistent with the training process of the impedance prediction model; the therapeutic effect is the degree of improvement in physiological function or relief of pathological condition shown by the user after receiving biocurrent therapy, such as pain relief, muscle relaxation, improved blood flow, etc.; the therapeutic effects corresponding to each particle are compared separately, and the particle with the largest therapeutic effect is marked as the optimal particle, and the particle with the smallest therapeutic effect is marked as the worst particle.

[0044] In the above step S605, the expression of the charge value is: ; Where, For the The charge value of a particle, is an exponential function, For the The therapeutic effect of particles, , To achieve the best therapeutic effect of particles, For the therapeutic effect of the worst particles, A very small positive number used to avoid the denominator being 0.

[0045] In the above step S606, the expression of the guiding electromagnetic force is: ; Where, For the The guiding electromagnetic force of particles, For the Particle pair The force exerted by each particle, and ; The expression of the acting force is: ; Where, For the particles, For the particles, For the Particles and The Euclidean distance between particles, For the The charge value of a particle.

[0046] In the above step S607, the expression for updating each particle is: ; Where, After the update particles, is the step size parameter, which controls the amplitude of each update and is set by those skilled in the art according to actual conditions. For the The magnitude of the electromagnetic force guiding each particle, For the The normalized direction vector of each particle is used to ensure the correct movement direction and the same step size for each update. For interval Random numbers within.

[0047] In the above step S608, the expression for locally updating the optimal particle is: ; Where, is the optimal particle after local update, is the optimal particle, is the local step size, is a random perturbation vector; both the local step size and the random perturbation vector are preset by those skilled in the art according to actual conditions.

[0048] In the above step S609, the iteration threshold is preset by those skilled in the art according to actual conditions.

[0049] It should be noted that when updating particles in a particle group, if there is a parameter in the parameter set corresponding to the updated particle that is greater than the maximum value of the corresponding value range, the value of the corresponding parameter will be corrected to the maximum value of the corresponding value range; if there is a parameter that is less than the minimum value of the corresponding value range, the value of the corresponding parameter will be corrected to the minimum value of the corresponding value range.

[0050] It should be understood that when the biocurrents acting on different bioelectrical impedance areas are collaboratively modulated based on the optimal configuration scheme of physical therapy parameters, multiple groups of EMS biocurrents are adjusted simultaneously through the same control chip, and the multiple groups of EMS biocurrents do not interfere with each other, thereby achieving multi-level and multi-frequency adjustment.

[0051] This embodiment constructs differentiated bioelectrical impedance areas by real-time acquisition of multi-dimensional human physiological state data, accurately depicting the tissue characteristics and electrophysiological environment of the target treatment area; based on in-depth analysis of human tissue sensitivity and tissue penetration, it dynamically evaluates the response characteristics of each bioelectrical impedance area to bioelectric current, and generates an energy conduction efficiency curve and a response type value matrix, providing a reliable basis for subsequent personalized regulation; adopts an intelligent optimization algorithm combined with treatment characteristic data to dynamically generate the optimal configuration scheme of physical therapy parameters, which can modulate bioelectric currents of different frequencies, intensities and other parameters for different bioelectrical impedance areas, and realize precise and coordinated multi-frequency treatment; realizes real-time perception and intelligent response to changes in physiological state during treatment, effectively solves the problems of poor targeting and rough regulation in traditional bioelectrical current treatment, and significantly improves the adaptability and application effect of bioelectric current in refined regulation and personalized treatment, thereby effectively improving the treatment effect and comfort.

[0052] Example 2 The present application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories may store computer-readable code that, when executed by the one or more processors, may execute the multi-frequency intelligent biocurrent coordinated modulation method described above.

[0053] The method or system according to the embodiment of the present application can also be implemented with the aid of the architecture of the electronic device shown in the present application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store a multi-frequency intelligent biocurrent collaborative modulation method provided in the present application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in the present application is only exemplary. When implementing different devices, one or more components in the electronic device shown in the present application may be omitted according to actual needs.

[0054] Example 3 As shown, one embodiment of the present application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, a multi-frequency intelligent biocurrent coordinated modulation method according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc.

[0055] Furthermore, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions capable of being executed by a processor to perform the instructions corresponding to the method steps provided herein, such as a multi-frequency intelligent biocurrent coordinated modulation method. When executed by a central processing unit (CPU), this computer program performs the aforementioned functions defined in the method of the present application.

[0056] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will be able to modify the technical solutions described in the foregoing embodiments or to substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

[0057] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0058] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0059] In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0060] In the description of the present invention, “several” means one or more, and “a large number” means two or more.

[0061] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0062] The formulas in this manual are all dimensionless and calculated using numerical values. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field based on actual conditions.

[0063] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

Claims

1. A multi-frequency intelligent biocurrent collaborative modulation method, characterized in that: include: Step S1: Real-time collection of human physiological status data of each sub-treatment area within the treatment area, where the human physiological status data includes skin physiological data, tissue structure data, muscle physiological data, and blood circulation data; Step S2: Fusing muscle physiological data with blood circulation data, performing bioelectrical impedance analysis on each sub-treatment area, performing connectivity analysis on all sub-treatment areas based on the bioelectrical impedance analysis results, and constructing differentiated bioelectrical impedance areas; Step S3: Based on the skin physiological data and tissue structure data, the sensitivity of human tissue in each bioelectrical impedance area is deeply characterized, and combined with the blood circulation data, the tissue penetration degree of each bioelectrical impedance area is comprehensively evaluated; Step S4: collecting treatment characteristic data in real time, and combining it with the degree of tissue penetration, extracting the energy conduction efficiency of each bioelectrical impedance area, and generating an energy conduction efficiency curve; Step S5: combining skin physiological data and blood circulation data to quantitatively evaluate the tissue response factors of each bioelectrical impedance area, and classifying and labeling the tissue response factors based on a threshold decision mechanism to generate a response type value matrix; Step S6: Integrate the energy conduction efficiency curve, the response type value matrix, and the treatment characteristic data to dynamically formulate an optimal configuration plan for the therapy parameters, and coordinately modulate the biocurrents acting on different bioelectrical impedance areas based on the optimal configuration plan for the therapy parameters.

2. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 1, characterized in that: The skin physiological data includes skin surface temperature, skin moisture content, skin pH value, sebum secretion and skin elasticity coefficient; the tissue structure data is subcutaneous fat thickness; the muscle physiological data is muscle tension; The blood circulation data includes blood flow rate and blood oxygen saturation; The method of performing bioelectrical impedance analysis on each sub-treatment area includes: The same muscle tension and blood flow rate in the corresponding sub-treatment area are used as a set of impedance analysis data, and the impedance analysis data corresponds to the sub-treatment area one by one; Each set of impedance analysis data is input into the trained impedance prediction model to predict the electrical impedance value corresponding to each sub-treatment area.

3. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 2, characterized in that: The method for constructing differentiated bioelectrical impedance regions comprises: A clustering algorithm is used to cluster the electrical impedance values of all sub-treatment areas to obtain two impedance categories; the electrical impedance values in the two impedance categories are averaged to obtain the electrical impedance mean; the two electrical impedance means are compared, and the impedance category with the larger electrical impedance mean is marked as the high-impedance category, and the impedance category with the smaller electrical impedance mean is marked as the low-impedance category; if the electrical impedance value is in the high-impedance category, the area category of the corresponding sub-treatment area is defined as the high-impedance area; if the electrical impedance value is in the low-impedance category, the area category of the corresponding sub-treatment area is defined as the low-impedance area; The adjacent areas corresponding to each sub-treatment area are taken as a set of adjacent sets, and the adjacent sets correspond to the sub-treatment areas one by one. In each set of adjacent sets, the adjacent areas with the same area category as the corresponding sub-treatment area are screened out and marked as connected areas. An associated edge is established between each connected area and the corresponding sub-treatment area, and all sub-treatment areas are taken as nodes. An associated graph is constructed based on the nodes and associated edges. In the associated graph, the depth-first search method is used to search for connected components and obtain Connected components are formed, and the sub-treatment areas corresponding to the nodes in each connected component are merged to construct a bioelectrical impedance area. The bioelectrical impedance area corresponds to the connected components one by one. is an integer greater than 0; if the region categories of all adjacent regions in the adjacent set are different from the corresponding sub-treatment region, the corresponding sub-treatment region will be regarded as a bioelectrical impedance region.

4. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 3, characterized in that: The method for in-depth characterization of the sensitivity of human tissue in each bioelectrical impedance region includes: Corresponding neural network encoder models are constructed for skin surface temperature, subcutaneous fat thickness and skin moisture content respectively; based on the neural network encoder model, the skin surface temperature, subcutaneous fat thickness and skin moisture content of each bioelectrical impedance area are mapped into corresponding high-dimensional feature vectors respectively; the same high-dimensional feature vectors of the corresponding bioelectrical impedance area are regarded as a group of feature sets, and each group of feature sets is input into the pre-constructed tensor fusion network in turn for a three-modal tensor product operation to generate a sensitivity fusion vector corresponding to each bioelectrical impedance area; each sensitivity fusion vector is sequentially subjected to dimensionality reduction operation through the pre-constructed fully connected neural network to obtain the sensitivity feature vector corresponding to each bioelectrical impedance area as an in-depth representation of human tissue sensitivity.

5. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 4, characterized in that: The method for comprehensively evaluating the tissue penetration degree of each bioelectrical impedance area includes: A corresponding neural network encoder model is constructed for blood oxygen saturation. Based on the neural network encoder model, the blood oxygen saturation of each bioelectrical impedance area is mapped to a corresponding blood oxygen feature vector. A weight set is preset, and the weight set includes weight coefficients corresponding to the sensitivity feature vector and the blood oxygen feature vector. Based on the weight set, the sensitivity feature vector and the blood oxygen feature vector of each bioelectrical impedance area are weightedly summed to obtain the tissue penetration degree of each bioelectrical impedance area.

6. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 5, characterized in that: The treatment characteristic data includes the area and depth information of each bioelectrical impedance area; The method for extracting the energy conduction efficiency of each bioelectrical impedance region includes: A corresponding tissue conductivity model and current density distribution model are constructed for each bioelectrical impedance region; the tissue conductivity model and current density distribution model corresponding to the bioelectrical impedance region are fused to obtain a fused simulation model; The electrical impedance value, area, depth information, and tissue penetration degree corresponding to each bioelectrical impedance region are input into the corresponding fusion simulation model, and multiple biocurrents are set as input for each fusion simulation model. Numerical simulation technology is used to simulate each fusion simulation model in turn, and the energy conduction ratio corresponding to each biocurrent is output. All energy conduction ratios with the same corresponding fusion simulation model are averaged to obtain the simulation efficiency of each bioelectrical impedance region. The area, depth information, and tissue penetration degree of each bioimpedance region are sequentially standardized to obtain standardized data for each bioimpedance region; the standardized data includes the standardized area, standard depth, and standard penetration degree; the analytical efficiency of each bioimpedance region is calculated based on the functional relationship between each type of data within the standardized data; the simulation ratio of each bioimpedance region is calculated based on the simulation efficiency, and the analytical ratio of each bioimpedance region is calculated based on the analytical efficiency; and the sum of the simulation ratio and analytical ratio for the same bioimpedance region is used as the energy conduction efficiency of the corresponding bioimpedance region; The method for generating an energy conduction efficiency curve comprises: The energy conduction efficiency of each bioelectrical impedance area is sorted in ascending order to generate an efficiency sequence; according to the positive order of the efficiency sequence, the bioelectrical impedance area corresponding to each energy conduction efficiency is set with a numerical label in ascending order and marked as a regional label; the regional label corresponding to each energy conduction efficiency is used as the horizontal axis index, and the corresponding energy conduction efficiency is used as the vertical axis value to generate an energy conduction efficiency curve.

7. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 6, characterized in that: The step of quantitatively evaluating the tissue response factor of each bioelectrical impedance area includes: Step S501: The skin pH value, sebum secretion, skin elasticity coefficient and blood flow rate of each bioelectrical impedance area are taken as a set of impact analysis data; Step S502: construct multiple fuzzy sets for each data in the impact analysis data; Step S503: converting each set of impact analysis data into the membership degree of each corresponding fuzzy set through fuzzification technology; Step S504: defining fuzzy rules; Step S505: Match each set of fuzzified impact analysis data with fuzzy rules respectively, and perform fuzzy reasoning using a fuzzy reasoning method to obtain the fuzzy reasoning results corresponding to each bioelectrical impedance area. The fuzzy reasoning results are the membership of each response level, which includes high level, medium level, and low level. Step S506: Set the factor interval. The range of the factor interval is ; Divide the factor interval evenly into three level intervals, and the level intervals correspond one-to-one to the levels in the response level; take the mean of the maximum and minimum values of each level interval as the interval mean of the corresponding level interval; Step S507: Based on the interval mean, perform weighted summation on the membership of each response level corresponding to each bioelectrical impedance area to obtain the total response value of each bioelectrical impedance area; add the membership of each response level corresponding to each bioelectrical impedance area in sequence to obtain the total membership; and use the ratio of the total response value of each bioelectrical impedance area to the corresponding total membership as the tissue response factor of each bioelectrical impedance area.

8. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 7, characterized in that: The method for generating a response type value matrix includes: According to the area type corresponding to the treatment area, the corresponding response threshold is obtained from a pre-constructed threshold set; the threshold set includes response thresholds corresponding to different area types, and each tissue response factor is compared with the response threshold respectively; the tissue response factor with a value greater than or equal to the response threshold is marked as a positive response factor, and the tissue response factor with a value less than the response threshold is marked as a negative response factor; different digital labels are set for the positive response factor and the negative response factor, and marked as response type values; according to the tissue response factor of each bioelectrical impedance area, the corresponding response type value is obtained, and a response type value matrix is generated based on the response type value.

9. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 8, characterized in that: The step of dynamically formulating the optimal configuration plan for physical therapy parameters includes: Step S601: Acquire parameter ranges based on the energy conduction efficiency curve, the response type value matrix, the treatment characteristic data, and the electrical impedance values of each bioelectrical impedance region; Step S602: Construct based on parameter range Group parameter collection, is an integer greater than 1; Step S603: From Randomly select from the group parameter set Group parameter set and All group parameter sets are treated as particles to construct a particle group, and the number of iterations is set to 0; Step S604: determining the best particle and the worst particle in the particle group; Step S605: Calculate the charge value corresponding to each particle in the particle group; Step S606: Calculating the guiding electromagnetic force of each particle in the particle group based on the charge value; Step S607: updating each particle in the particle group according to the guiding electromagnetic force; Step S608: locally updating the optimal particle in the particle group; Step S609: Compare the number of iterations with a preset iteration threshold. If the number of iterations is less than the iteration threshold, increment the number of iterations and return to step S604. If the number of iterations is greater than or equal to the iteration threshold, proceed to step S610. Step S610: obtaining the optimal particle in the particle group, marking the parameter set corresponding to the optimal particle as the optimal set, and using the optimal set as the optimal configuration scheme for the physical therapy parameters.

10. The multi-frequency intelligent biocurrent coordinated modulation method according to claim 9, characterized in that: In step S601, the parameter range includes the value range corresponding to each parameter in the physical therapy parameters; The method for obtaining the parameter range is as follows: the energy conduction efficiency curve, the response type value matrix, the treatment characteristic data, and the electrical impedance values of each bioelectrical impedance region are used as range analysis data, and the range analysis data are input into the trained range prediction model to predict the corresponding parameter range; In step S602, construct The method of grouping parameter sets is: randomly select a value from each value range within the parameter range and construct a set of parameter sets; and so on. Group parameter collection, The set of group parameters are all different; In step S604, the method for determining the best particle and the worst particle in the particle group includes: The range analysis data and the parameter set corresponding to each particle are regarded as a set of effect analysis data, and the effect analysis data and the parameter set correspond one-to-one; each set of effect analysis data is input into the trained effect prediction model to predict the physical therapy effect corresponding to each set of effect analysis data; the physical therapy effect corresponding to each particle is compared respectively, and the particle with the largest physical therapy effect is marked as the optimal particle, and the particle with the smallest physical therapy effect is marked as the worst particle.

Images