Magnetic therapy bed optimization control method and system based on brain wave driving

Through the optimization control method of magnetic therapy bed based on brain wave-driven, the magnetic field parameters are dynamically adjusted using machine learning and fuzzy control algorithms, and the problem that existing magnetic therapy equipment cannot be personalized is solved, achieving more efficient magnetic therapy effects and user experience .

CN118938719APending Publication Date: 2024-11-12HEYE HEALTH TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410857519.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-28
Publication Date
2024-11-12

Smart Images

  • Figure CN118938719A_ABST
    Figure CN118938719A_ABST
Patent Text Reader

Abstract

The invention provides a magnetic therapy bed optimization control method and system based on brain wave driving, and relates to the technical field of smart home, and the method comprises the steps: obtaining the physiological data and environmental parameter data of a magnetic therapy bed user, and obtaining a standardized data set; utilizing a machine learning algorithm and an adaptive learning algorithm to obtain specific parameters of physiological and psychological states of the user and optimized magnetic field parameters; performing brain dynamic trajectory prediction by using time sequence analysis and a recurrent neural network, and performing adaptive learning algorithm processing to obtain an optimal dynamic magnetic field parameter; a personalized optimized magnetic field parameter is obtained; a dynamic map is established through brain wave feature extraction and a dynamic time warping algorithm, and optimal control over the magnetic therapy bed is achieved. According to the invention, high precision, real-time optimization and individuation of control of the magnetic therapy bed are realized, the magnetic therapy effect and the user experience are remarkably improved, and the magnetic therapy bed has important innovativeness and practical value.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to a brainwave-driven magnetic therapy bed optimization control method and system. Background Art

[0002] As a non-invasive, drug-free treatment method, magnetic therapy has been widely used in the medical and health care fields. Its main principle is to affect the physiological activities of the human body through magnetic fields to achieve the purpose of treating and alleviating symptoms. Existing magnetic therapy devices mostly use fixed magnetic fields or simple adjustable magnetic fields for treatment, but these devices generally have the following problems: First, fixed magnetic fields cannot provide personalized treatment for individual differences and specific symptoms of different users, and the effect is limited; second, simple adjustable magnetic fields lack real-time feedback mechanisms and cannot be dynamically adjusted according to the user's real-time physiological state, resulting in poor treatment effects; finally, existing magnetic therapy devices usually rely on manual adjustments and experience settings, making it difficult to achieve precise control and optimize treatment effects.

[0003] In order to solve these problems, the existing technology has proposed some improvement methods, such as adding sensors to monitor the user's physiological data and using simple control algorithms to adjust the magnetic field parameters. However, these methods still have the following shortcomings: the types and accuracy of monitoring data are limited, the control algorithm is simple and the response speed is slow, and it is difficult to achieve true real-time optimization and personalized treatment. Therefore, the existing technology still has a lot of room for improvement in improving the effect of magnetic therapy and user experience. Summary of the invention

[0004] The purpose of the present invention is to provide a brainwave-driven magnetic therapy bed optimization control method and system to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a method for optimizing and controlling a magnetic therapy bed based on brainwave drive, comprising:

[0006] Acquire the physiological data and environmental parameter data of the magnetic therapy bed user, including environmental status data, brain wave data, and electric field parameter data, and obtain a standardized data set after data preprocessing and normalization;

[0007] Based on standardized data sets, biofeedback analysis is performed using machine learning algorithms and adaptive learning algorithms. After state analysis and parameter optimization, the specific parameters of the user's physiological and psychological state and the optimized magnetic field parameters are obtained.

[0008] The magnetic field parameters are combined with the user's real-time brain wave data, and the dynamic trajectory of the brain is predicted using time series analysis and recursive neural networks. After being processed by an adaptive learning algorithm, the intensity and frequency of the magnetic field are dynamically adjusted to obtain the optimal dynamic magnetic field parameters.

[0009] According to the optimal dynamic magnetic field parameters, the magnetic field parameters are further adjusted using the fuzzy control algorithm to obtain personalized optimized magnetic field parameters;

[0010] Multi-dimensional data fusion technology is used to fuse the personalized optimized magnetic field parameters. A dynamic map is established through brain wave feature extraction and dynamic time warping algorithm. Combined with the data collected in real time by the electric field sensor, the optimal electric field strength and frequency are calculated through finite element analysis and ant colony optimization algorithm. Through the real-time feedback system, the magnetic field parameters are dynamically adjusted to achieve optimized control of the magnetic therapy bed.

[0011] Preferably, the biofeedback analysis is performed based on the standardized data set using a machine learning algorithm and an adaptive learning algorithm, and after state analysis and parameter optimization, specific parameters of the user's physiological and psychological state and optimized magnetic field parameters are obtained, including:

[0012] According to the environmental state data, brain wave data and electric field parameter data, the autoregressive model is used to extract the time series features to obtain the autoregressive coefficients, and each time series data is subjected to wavelet transformation to decompose it into different frequency components, and the wavelet coefficients are extracted as features. The time series data is subjected to Fourier transformation to obtain frequency domain features and form a high-dimensional feature vector;

[0013] Based on the high-dimensional feature vector, a multi-layer perceptron neural network model is constructed, and the back-propagation algorithm and gradient descent optimizer are used to train the neural network, adjust the weights and biases, and after the training is completed, the new feature vector is input, and the neural network outputs the predicted user's physiological and psychological state to obtain the classification result, which includes the user's physiological and psychological state;

[0014] The classification results are processed by an adaptive learning algorithm, the magnetic field parameters are adjusted, the reward value is set according to the user's status feedback, and the reinforcement learning algorithm is used to optimize the strategy. The optimal action is selected in each state to maximize the cumulative reward. The training is repeated in a simulated environment, and the Q value is updated until convergence. Based on the trained Q value function, the optimal magnetic field parameters in the current state are selected, and the user's feedback includes treatment effect and comfort.

[0015] Preferably, the magnetic field parameters are combined with the user's real-time brain wave data, and the brain dynamic trajectory is predicted using time series analysis and recursive neural network, and the intensity and frequency of the magnetic field are dynamically adjusted through adaptive learning algorithm processing to obtain the optimal dynamic magnetic field parameters, including:

[0016] The user's real-time brain wave data is obtained, and the magnetic field parameters and the user's real-time brain wave data are synchronized to a unified time axis to obtain combined data. The calculation formula is as follows:

[0017] D combined (t) = [EEG (t), B (t)]

[0018] Where, EEG(t) represents the brain wave data at time point t, and B(t) represents the magnetic field parameters at time point t;

[0019] Based on the combined data, the autoregressive model is used to model and extract features of the time series to obtain the characteristics of the brain dynamic trajectory. The characteristic vector is obtained through the coefficients and prediction residuals of the AR model.

[0020] According to the extracted feature vectors, the future brain dynamic trajectory is predicted through recursive neural network processing to obtain the predicted brain wave data;

[0021] Based on the predicted brain wave data, after reinforcement learning and deep Q network, the magnetic field parameters are continuously adjusted according to the reward mechanism to find the optimal strategy, and finally the optimal dynamic magnetic field parameters are obtained. The calculation formula is as follows:

[0022]

[0023] Where Q(s,a) is the value of taking action a in state s, α is the learning rate, r is the reward value, γ is the discount factor, s′ and a′ are the next state and action.

[0024] Preferably, the magnetic field parameters are further adjusted using a fuzzy control algorithm based on the optimal dynamic magnetic field parameters to obtain personalized optimized magnetic field parameters, which include:

[0025] The user's real-time brain wave data and optimal dynamic magnetic field parameters are input into the fuzzy logic system, and the input data is fuzzified through the membership function to obtain the fuzzified input data;

[0026] According to the fuzzified input data, fuzzy output data is obtained through fuzzy reasoning and fuzzy rule base processing, and based on the fuzzy output data, the adjusted magnetic field parameters are obtained through defuzzification processing.

[0027] Preferably, the multi-dimensional data fusion technology is used to fuse the personalized optimized magnetic field parameters, and a dynamic spectrum is established through brain wave feature extraction and dynamic time warping algorithm, which includes:

[0028] The personalized optimized magnetic field parameters and the acquired real-time brain wave data of the user are processed by multi-dimensional data fusion technology, the magnetic field parameters and brain wave data are aligned to a unified time axis, and the data of different dimensions are fused into a unified multi-dimensional data set through clustering analysis method. The clustering process includes: selecting an appropriate number of clusters k, clustering the data, randomly selecting the initial cluster center, iteratively updating the clusters and cluster centers to which the data points belong until convergence, and obtaining the fused multi-dimensional data set. The calculation formula is as follows:

[0029]

[0030] In the formula, J is the clustering objective function, k is the number of clusters, x is the data point, and C i is the i-th cluster, u i is the center of the i-th cluster;

[0031] Based on the fusion data set, the wavelet transform method is used to extract key features from the EEG data to obtain EEG features, wherein the EEG signal is decomposed into different frequency bands, the time-frequency features are extracted, and the time series data containing feature information is obtained;

[0032] The EEG features are used as the input data of dynamic time warping. By aligning the time series, the minimum distance between the two time series is calculated to obtain the aligned time series.

[0033] The aligned time series data is embedded into the high-dimensional phase space, and the appropriate delay time and embedding dimension are selected to convert the time series into trajectories in the phase space. The deep learning algorithm is used to extract the spatial features in the space and the temporal features in the time series for fusion, thereby constructing a dynamic graph, which includes the connection relationship between different nodes and edges, as well as the magnetic field parameters used to quantify each user's position.

[0034] Preferably, the data collected in real time by the electric field sensor is combined, and the optimal electric field strength and frequency are calculated through finite element analysis and ant colony optimization algorithm. The magnetic field parameters are dynamically adjusted through a real-time feedback system to achieve optimal control of the magnetic therapy bed, including:

[0035] Using brain wave characteristics, pressure, temperature and magnetic field distribution data in dynamic graphs, a three-dimensional finite element model of the magnetic therapy bed was established to determine boundary conditions and load conditions, including electric field strength, frequency and user's physiological parameters;

[0036] The finite element method is used to solve the electric field distribution, and the electric field strength and frequency distribution at each position are calculated; based on the electric field strength and frequency distribution, the ant colony optimization algorithm is used to obtain the optimal electric field strength and optimal frequency parameters, wherein the processing process includes each ant selecting a path according to the pheromone concentration and the heuristic function, constructing candidate solutions for the electric field strength and frequency, updating the pheromone concentration on the path according to the objective function value found by each ant, and repeating the path selection and pheromone update process until the stop condition is met;

[0037] The optimal electric field strength and optimal frequency parameters are input into the real-time feedback system. The real-time feedback system monitors the user's physiological data, dynamically adjusts the magnetic field parameters based on the optimization results, and uses a fuzzy control algorithm for fine-tuning, thereby optimizing the control of the magnetic therapy bed.

[0038] In the second aspect, the present application also provides a brainwave-driven magnetic therapy bed optimization control system, comprising:

[0039] Acquisition module: used to obtain the physiological data and environmental parameter data of the magnetic therapy bed user, including environmental status data, brain wave data and electric field parameter data, and obtain a standardized data set after data preprocessing and normalization;

[0040] Optimization module: used to perform biofeedback analysis based on standardized data sets, using machine learning algorithms and adaptive learning algorithms. After state analysis and parameter optimization, the specific parameters of the user's physiological and psychological state and the optimized magnetic field parameters are obtained;

[0041] Processing module: used to combine magnetic field parameters with the user's real-time brain wave data, use time series analysis and recursive neural network to predict brain dynamic trajectory, and dynamically adjust the intensity and frequency of the magnetic field through adaptive learning algorithm processing to obtain the optimal dynamic magnetic field parameters;

[0042] Adjustment module: used to further adjust the magnetic field parameters according to the optimal dynamic magnetic field parameters using fuzzy control algorithm to obtain personalized optimized magnetic field parameters;

[0043] Control module: It is used to fuse the personalized optimized magnetic field parameters using multi-dimensional data fusion technology, establish a dynamic map through brain wave feature extraction and dynamic time warping algorithm, and combine the data collected in real time by the electric field sensor. Through finite element analysis and ant colony optimization algorithm, the optimal electric field strength and frequency are calculated, and the magnetic field parameters are dynamically adjusted through the real-time feedback system to achieve optimized control of the magnetic therapy bed.

[0044] The beneficial effects of the present invention are:

[0045] The present invention uses a high-density EEG headset and various sensors to obtain the user's brain wave data, environmental status data and electric field parameter data in real time, and generates a standardized data set after preprocessing and normalization, ensuring the accuracy and consistency of the data, providing a reliable basis for subsequent analysis and processing, and improving the overall efficiency and accuracy of the system.

[0046] The present invention utilizes machine learning algorithms and adaptive learning algorithms to perform biofeedback analysis on standardized data sets, including state analysis and parameter optimization, combines magnetic field parameters with real-time brain wave data of users, utilizes time series analysis and recursive neural networks to predict brain dynamic trajectories, and dynamically adjusts the intensity and frequency of the magnetic field through processing by the adaptive learning algorithm to obtain optimal dynamic magnetic field parameters. The present invention can quickly and accurately identify the physiological and psychological state of the user, and optimize the magnetic field parameters in real time, making the magnetic therapy effect more significant and personalized, realizing real-time optimization of the magnetic field parameters, so that the magnetic therapy bed can be dynamically adjusted according to the real-time state of the user, thereby improving the accuracy and timeliness of the treatment effect.

[0047] The present invention can process complex and ambiguous data, improve the flexibility and adaptability of the system, ensure that different users can obtain the best personalized treatment effect, adopt multi-dimensional data fusion technology, extract and cluster analysis of brain wave features, and establish a dynamic map through dynamic time warping algorithm, combine the data collected in real time by the electric field sensor, calculate the optimal electric field strength and frequency through finite element analysis and ant colony optimization algorithm, and provide quantitative analysis results of pressure, temperature and magnetic field distribution at various positions of the user, so that the control of the magnetic therapy bed is more scientific and accurate, and the treatment plan is optimized.

[0048] The present invention inputs the optimal electric field strength and frequency parameters into the real-time feedback system, dynamically adjusts the magnetic field parameters in combination with the user's physiological data, and uses a fuzzy control algorithm for fine-tuning, thereby realizing real-time optimization control of the magnetic therapy bed, ensuring the effectiveness and safety of magnetic field regulation, and improving the user's treatment experience.

[0049] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 It is a schematic flow chart of the optimization control method of the magnetic therapy bed based on brain wave drive described in an embodiment of the present invention;

[0052] Figure 2 Schematic diagram of the structure of the brainwave-driven magnetic therapy bed optimization control system described in an embodiment of the present invention.

[0053] In the figure: 701, acquisition module; 702, optimization module; 7021, decomposition and transformation unit; 7022, training unit; 7023, selection unit; 703, processing module; 7031, acquisition unit; 7032, feature extraction unit; 7033, prediction unit; 7034, acquisition unit; 704, adjustment module; 7041, input unit; 7042, fuzzy processing unit; 705, control module; 7051, clustering processing unit; 7052, feature extraction unit; 7053, construction unit; 7054, establishment unit; 7055, solution unit; 7056, control unit. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0055] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0056] Embodiment 1:

[0057] This embodiment provides a method for optimizing and controlling a magnetic therapy bed based on brainwave drive.

[0058] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400 and step S500.

[0059] S100, obtaining physiological data and environmental parameter data of the user of the magnetic therapy bed, including environmental state data, brain wave data, and electric field parameter data, and obtaining a standardized data set through data preprocessing and normalization.

[0060] It can be understood that in this step, the data obtained include environmental status data, brain wave data and electric field parameter data, wherein the environmental status data includes room temperature, humidity, light intensity, etc., which can be obtained through environmental sensors integrated in the magnetic therapy bed, for example, the temperature sensor can detect the temperature of the room, the humidity sensor can detect the humidity of the air, and the light sensor can detect the light intensity of the room; the brain wave data includes but is not limited to the use of high-density EEG equipment (such as 64-channel or 128-channel EEG caps) to collect the user's brain wave data in real time. The EEG device can accurately capture the electrical activity of different areas of the brain, and the data usually exists in the form of a multi-channel time series; the electric field parameter data includes the intensity and frequency of the electric field, which can be monitored in real time by the electric field sensor integrated in the magnetic therapy bed. The electric field parameters are an important factor in magnetic therapy and have a direct impact on the user's treatment effect.

[0061] In this step, specifically, the environmental state data, brain wave data and electric field parameter data are denoised to remove interference signals and retain valid data. Common denoising methods include low-pass filtering, high-pass filtering, band-pass filtering and adaptive filtering, etc., wherein missing values ​​in the data are filled, and commonly used methods include mean filling, interpolation and multiple interpolation, etc. If the environmental state data, brain wave data and electric field parameter data are normalized by a normalization method, the data are mapped to a unified range (such as between 0 and 1); commonly used normalization methods include Min-Max normalization, Z-score standardization, etc. Finally, the pre-processed and normalized environmental state data, brain wave data and electric field parameter data are combined into a standardized data set, which can be regarded as a multidimensional data matrix, each row represents multidimensional data at a time point, and each column represents a feature.

[0062] In summary, through data preprocessing and denoising, noise and interference are removed to ensure the accuracy and reliability of the data; through time synchronization and missing value filling, the consistency of data from different sources at the same time point is ensured to improve the integrity of the data; through normalization, data of different scales are mapped to a unified range to facilitate subsequent feature extraction and algorithm application; a standardized data set is obtained, which provides a high-quality data foundation for subsequent biofeedback analysis and parameter optimization.

[0063] S200, based on standardized data sets, using machine learning algorithms and adaptive learning algorithms, conduct biofeedback analysis, and obtain specific parameters of the user's physiological and psychological states and optimized magnetic field parameters through state analysis and parameter optimization.

[0064] It can be understood that step S200 includes S201, S202 and S203, wherein:

[0065] S201, according to the environmental state data, brain wave data and electric field parameter data, using the autoregressive model to extract time series features, obtain autoregressive coefficients, and perform wavelet transform on each time series data to decompose it into different frequency components, extract the wavelet coefficients as features, and perform Fourier transform on the time series data to obtain frequency domain features to form a high-dimensional feature vector;

[0066] It can be understood that feature extraction ensures the comprehensive acquisition of the time-frequency characteristics and time series characteristics of the data, laying a solid foundation for the subsequent state analysis, where the feature vectors, including time-frequency domain features and time series features, represent the user's brainwave patterns and environmental states.

[0067] S202, based on the high-dimensional feature vector, construct a multi-layer perceptron neural network model, and use the back propagation algorithm and the gradient descent optimizer to train the neural network, adjust the weights and biases, and after the training is completed, input the new feature vector, and the neural network outputs the predicted physiological and psychological state of the user to obtain a classification result, which includes the physiological state and psychological state of the user;

[0068] Among them, the classification results include the user's physiological state (such as attention level, fatigue level) and psychological state (such as emotional state). Through the application of classification algorithms, accurate identification of the user's physiological and psychological state is achieved, providing a basis for personalized optimization of magnetic field parameters.

[0069] S203. The classification results are processed by an adaptive learning algorithm, the magnetic field parameters are adjusted, the reward value is set according to the feedback of the user status, the reinforcement learning algorithm is used to optimize the strategy, the optimal action is selected in each state to maximize the cumulative reward, the training is repeated in a simulation environment, the Q value is updated until convergence, and the optimal magnetic field parameters in the current state are selected according to the trained Q value function, wherein the user feedback includes the treatment effect and comfort.

[0070] It should be noted that in this step, the magnetic field parameters are dynamically adjusted using an adaptive learning algorithm, which improves the personalized treatment effect of the magnetic therapy bed. The magnetic field parameters are continuously optimized through a real-time feedback system to ensure that the magnetic field environment during the treatment process is always in the best state, thereby improving the effectiveness of the treatment and user experience. By combining machine learning and adaptive learning algorithms, the user's physiological and psychological state can be dynamically identified, thereby achieving personalized treatment plans and improving treatment effects.

[0071] S300, combining magnetic field parameters with the user's real-time brain wave data, using time series analysis and recursive neural network to predict brain dynamic trajectory, and through adaptive learning algorithm processing, dynamically adjusting the intensity and frequency of the magnetic field to obtain the optimal dynamic magnetic field parameters.

[0072] It can be understood that step S300 includes S301, S302, S303 and S304, wherein:

[0073] S301, obtaining the user's real-time brain wave data, synchronizing the magnetic field parameters and the user's real-time brain wave data to a unified time axis, and obtaining combined data, the calculation formula of which is as follows:

[0074] D combined (t) = [EEG (t), B (t)]

[0075] Where, EEG(t) represents the brain wave data at time point t, and B(t) represents the magnetic field parameters at time point t;

[0076] S302. Based on the combined data, an autoregressive model is used to model and extract features of the time series to obtain the features of the brain dynamic trajectory, and a feature vector is obtained through the coefficients and prediction residuals of the AR model.

[0077] It should be noted that a high-density EEG device (such as a 64-channel or 128-channel EEG cap) is used to monitor the user's brain wave signals in real time. The EEG device can capture the electrical activity of different areas of the brain and provide high-resolution time series data. The collected brain wave data is transmitted to the data processing system in real time by wireless or wired transmission. The integrity and low latency of the data need to be guaranteed during the transmission process. In this step, since the acquisition frequency and timestamp of brain wave data and magnetic field parameter data may be different, it is necessary to synchronize these data, and a common method is interpolation and timestamp alignment. Among them, using methods such as linear interpolation or spline interpolation, data of different frequencies are synchronized to a unified time axis, and the timestamps of brain wave data and magnetic field parameter data are aligned to ensure that the data of both at the same time point can correspond.

[0078] The synchronized brain wave data and magnetic field parameter data are combined into a multidimensional data set. The data at each time point includes EEG signals and corresponding magnetic field parameters. Through timeline synchronization, the temporal consistency of brain wave data and magnetic field parameter data is ensured, providing accurate basic data for subsequent analysis and processing, realizing real-time monitoring and synchronous processing of user brain waves and magnetic field parameters, and facilitating instant response and adjustment of magnetic therapy parameters.

[0079] It should be noted that in step S302, the autoregressive model is a time series modeling method used to describe the inherent laws of time series data, in which the AR model assumes that the current time series value can be predicted by the values ​​of the previous time points. The model is trained using combined data, and the prediction residual of the AR model is calculated, that is, the difference between the actual value and the model predicted value. The autoregressive coefficient and the prediction residual of the AR model are combined into a feature vector F, which is used to describe the characteristics of the brain dynamic trajectory. The feature vector of the brain dynamic trajectory is extracted through the autoregressive model, which effectively describes the inherent laws and changing trends of the time series, improves the prediction accuracy of brain wave data, and compresses the high-dimensional time series data into a low-dimensional feature vector in the feature extraction process, reduces the computational complexity, and facilitates subsequent classification and optimization.

[0080] S303, predicting the future brain dynamic trajectory based on the extracted feature vector through recursive neural network processing to obtain predicted brain wave data;

[0081] S304. Based on the predicted brain wave data, after reinforcement learning and deep Q network, the magnetic field parameters are continuously adjusted according to the reward mechanism to find the optimal strategy, and finally the optimal dynamic magnetic field parameters are obtained. The calculation formula is as follows:

[0082]

[0083] Where Q(s,a) is the value of taking action a in state s, α is the learning rate, r is the reward value, γ is the discount factor, s′ and a′ are the next state and action.

[0084] It should be noted that through time series analysis and recursive neural networks, the user's brain dynamic trajectory can be accurately predicted and the magnetic field parameters can be adjusted in advance; using adaptive learning algorithms, the intensity and frequency of the magnetic field can be dynamically adjusted in real time to ensure that the parameters during magnetic therapy are always in the optimal state; through feature extraction and trajectory prediction, personalized adjustments can be made based on each user's real-time brain wave data to improve the personalization and accuracy of the magnetic therapy effect.

[0085] S400. According to the optimal dynamic magnetic field parameters, the magnetic field parameters are further adjusted using a fuzzy control algorithm to adapt to the personalized needs and treatment effects of different users, thereby obtaining personalized optimized magnetic field parameters.

[0086] It can be understood that step S400 includes S401 and S402, wherein:

[0087] S401, inputting the user's real-time brain wave data and the optimal dynamic magnetic field parameters into the fuzzy logic system, and fuzzifying the input data through the membership function to obtain the fuzzified input data;

[0088] It should be noted that fuzzy logic can handle the uncertainty and ambiguity in brain wave data and magnetic field parameters, providing a flexible control method. Through fuzzy processing, it can perform fine-grained control on input data of different degrees and improve control accuracy.

[0089] Specifically, in this step, fuzzy sets of brain wave data and magnetic field parameters are defined. For example, brain wave data are divided into three fuzzy sets of "low", "medium" and "high", and magnetic field parameters are divided into three fuzzy sets of "weak", "moderate" and "strong". The membership function of each fuzzy set is defined to describe the degree to which the data belongs to each fuzzy set. Commonly used membership functions include triangle, trapezoid and Gaussian functions. During the processing, the user's real-time brain wave data and the optimal dynamic magnetic field parameters are input into the fuzzy logic system, and the input data is fuzzified by the membership function to obtain the fuzzified input data.

[0090] S402, according to the fuzzified input data, fuzzy reasoning and fuzzy rule base processing are performed to obtain fuzzy output data, and based on the fuzzy output data, adjusted magnetic field parameters are obtained through defuzzification processing.

[0091] It should be noted that commonly used defuzzification methods include the centroid method, the maximum membership method and the weighted average method. According to the membership function of the fuzzy output data, the centroid of the output data is calculated, and the defuzzified output value is used as the new magnetic field parameter to further adjust the intensity and frequency of the magnetic field. According to the defuzzified magnetic field parameters, the control system of the magnetic therapy equipment is adjusted to achieve the effect of optimized control.

[0092] It can be understood that through defuzzification processing, fuzzy output data can be converted into accurate control parameters, improving the accuracy of magnetic field adjustment. The defuzzification processing is fast, and the fuzzy reasoning results are applied to magnetic field control in real time, thereby improving the response speed of the system.

[0093] S500 uses multi-dimensional data fusion technology to fuse the personalized optimized magnetic field parameters, establishes a dynamic map through brain wave feature extraction and dynamic time warping algorithm, and combines the data collected in real time by the electric field sensor. After finite element analysis and ant colony optimization algorithm, the optimal electric field strength and frequency are calculated, and the magnetic field parameters are dynamically adjusted through the real-time feedback system to achieve optimized control of the magnetic therapy bed.

[0094] It can be understood that step S500 includes S501, S502 and S503, wherein:

[0095] S501, the personalized optimized magnetic field parameters and the acquired real-time brain wave data of the user are processed by multi-dimensional data fusion technology, the magnetic field parameters and the brain wave data are aligned to a unified time axis, and the data of different dimensions are fused into a unified multi-dimensional data set by a clustering analysis method, wherein the clustering process includes: selecting an appropriate number of clusters k, clustering the data, randomly selecting an initial cluster center, iteratively updating the clusters and cluster centers to which the data points belong, until convergence, to obtain a fused multi-dimensional data set, and the calculation formula is as follows:

[0096]

[0097] In the formula, J is the clustering objective function, k is the number of clusters, x is the data point, and C i is the i-th cluster, u i is the center of the i-th cluster;

[0098] It should be noted that in addition to clustering algorithms, data of different dimensions can also be fused into a unified multidimensional data set through methods such as weighted averaging and principal component analysis (PCA). Among them, weighted averaging: weighted averaging of magnetic field parameters and brain wave data, and the weight can be determined according to the importance and reliability of the data; principal component analysis (PCA): PCA analysis is performed on the fused data to extract the main features, reduce the data dimension, and improve the calculation efficiency.

[0099] Therefore, multidimensional data fusion technology integrates different types of data into a unified framework to facilitate comprehensive analysis and processing; and through fusion processing, it can reveal the correlation between different data sources and improve the accuracy and reliability of data analysis.

[0100] S502, based on the fused data set, using the wavelet transform method to extract key features from the brain wave data to obtain brain wave features, wherein the brain wave signal is decomposed into different frequency bands, time-frequency features are extracted, and time series data containing feature information is obtained;

[0101] It should be noted that through EEG feature extraction, key information in EEG data can be captured to improve the accuracy of subsequent analysis. In addition, the dynamic time warping algorithm can align EEG data at different times to ensure data consistency and comparability, and improve the reliability of analysis results.

[0102] Specifically, in this step, EEG feature extraction and DTW are two consecutive processing steps, but they have different functions. EEG feature extraction is used to obtain the key features of brain waves, while DTW is used to align time series data for consistency comparison in time. Among them, the data after EEG feature extraction is still time series data, but it contains feature information. The input data of DTW is the EEG time series data after feature extraction. By aligning these time series, the changes in brain waves at different time points can be better compared and analyzed, especially for the case of nonlinear time changes.

[0103] In summary, EEG feature extraction provides time series data containing feature information, which is then input into DTW for time alignment to ensure consistent comparison and analysis in time, thereby achieving more accurate EEG dynamic analysis.

[0104] S503. Using brain wave features as input data for dynamic time warping, aligning time series, calculating the minimum distance between two time series, and obtaining an aligned time series; embedding the data of the aligned time series into a high-dimensional phase space, selecting appropriate delay time and embedding dimension, converting the time series into a trajectory in the phase space, and using a deep learning algorithm to extract spatial features in the space and temporal features in the time series for fusion, thereby constructing a dynamic graph, wherein the dynamic graph includes connection relationships between different nodes and edges and magnetic field parameters for quantifying each position of the user.

[0105] Therefore, it can be understood that through the above steps, a detailed description is given of how to use multidimensional data fusion technology, EEG feature extraction and dynamic time warping algorithm to build a dynamic map. Each step contains specific algorithms and data processing methods, showing the specific implementation of data fusion, feature extraction and dynamic map construction.

[0106] It can be understood that step S500 includes S504, S505 and S506, wherein:

[0107] S504, using brain wave characteristics, pressure, temperature and magnetic field distribution data in the dynamic spectrum, establish a three-dimensional finite element model of the magnetic therapy bed, and determine boundary conditions and load conditions, including electric field strength, frequency and user's physiological parameters;

[0108] It should be noted that the finite element method is used to solve the electric field distribution. The finite element method divides the entire computational domain into a finite number of small units for analysis, and a simple interpolation function is used in each unit to approximate the distribution of the variable. It is necessary to set the potential function as φ, and use the Poisson equation to obtain a set of algebraic equations through finite element discretization, and solve these equations to obtain the electric field distribution.

[0109] S505, using the finite element method to solve the electric field distribution, and calculating the electric field strength and frequency distribution at each position; based on the electric field strength and frequency distribution, the ant colony optimization algorithm is used to obtain the optimal electric field strength and optimal frequency parameters, wherein the processing process includes each ant selecting a path according to the pheromone concentration and the heuristic function, constructing candidate solutions for the electric field strength and frequency, updating the pheromone concentration on the path according to the objective function value found by each ant, and repeating the path selection and pheromone update process until the stop condition is met;

[0110] It should be noted that this step includes constructing a finite element model: applying the personalized optimized magnetic field parameters obtained by multidimensional data fusion technology to the finite element model, and the model includes the three-dimensional geometric structure, electromagnetic material properties and boundary conditions of the magnetic therapy bed; discretization: discretizing the entire calculation area into a finite number of small units, and using the interpolation function to approximate the potential distribution within each unit.

[0111] S506. Input the optimal electric field strength and optimal frequency parameters into the real-time feedback system. The real-time feedback system monitors the user's physiological data, dynamically adjusts the magnetic field parameters based on the optimization results, and uses the fuzzy control algorithm for fine-tuning, thereby optimizing the control of the magnetic therapy bed.

[0112] It should be noted that in this step, fuzzy rule design: design fuzzy control rules based on expert experience and experimental data to fuzzify the relationship between the user's physiological state and magnetic field parameters; fuzzy reasoning: use the fuzzy logic reasoning engine to perform fuzzy reasoning based on real-time feedback data to adjust the magnetic field parameters; defuzzification: defuzzify the results of fuzzy reasoning, convert them into specific control commands, and fine-tune the magnetic field parameters.

[0113] In summary, through the above two steps, combined with the real-time feedback system and fuzzy control algorithm, the optimal control of the magnetic therapy bed is achieved, ensuring that the magnetic field parameters match the user's personalized needs and physiological state.

[0114] Embodiment 2:

[0115] like Figure 2 As shown, this embodiment provides an optimization control system for a magnetic therapy bed based on brainwave drive, see Figure 2 The system comprises:

[0116] Acquisition module 701: used to acquire physiological data and environmental parameter data of the user of the magnetic therapy bed, including environmental state data, brain wave data and electric field parameter data, and obtain a standardized data set after data preprocessing and normalization;

[0117] Optimization module 702: used to perform biofeedback analysis based on standardized data sets using machine learning algorithms and adaptive learning algorithms, and obtain specific parameters of the user's physiological and psychological states and optimized magnetic field parameters through state analysis and parameter optimization;

[0118] Processing module 703: used to combine magnetic field parameters with the user's real-time brain wave data, use time series analysis and recursive neural network to predict brain dynamic trajectory, and dynamically adjust the intensity and frequency of the magnetic field through adaptive learning algorithm processing to obtain the optimal dynamic magnetic field parameters;

[0119] Adjustment module 704: used to further adjust the magnetic field parameters according to the optimal dynamic magnetic field parameters using a fuzzy control algorithm to obtain personalized optimized magnetic field parameters;

[0120] Control module 705: used to use multi-dimensional data fusion technology to fuse the personalized optimized magnetic field parameters, establish a dynamic map through brain wave feature extraction and dynamic time warping algorithm, and combine the data collected in real time by the electric field sensor, calculate the optimal electric field strength and frequency through finite element analysis and ant colony optimization algorithm, and dynamically adjust the magnetic field parameters through the real-time feedback system to achieve optimized control of the magnetic therapy bed.

[0121] Specifically, the optimization module 702 includes:

[0122] Decomposition and transformation unit 7021: used to extract time series features using an autoregressive model according to environmental state data, brain wave data and electric field parameter data, obtain autoregressive coefficients, perform wavelet transformation on each time series data, decompose it into different frequency components, extract wavelet coefficients as features, perform Fourier transformation on the time series data, obtain frequency domain features, and form a high-dimensional feature vector;

[0123] Training unit 7022: used to construct a multi-layer perceptron neural network model based on the high-dimensional feature vector, and use the back-propagation algorithm and the gradient descent optimizer to train the neural network, adjust the weights and biases, and after the training is completed, input the new feature vector, and the neural network outputs the predicted physiological and psychological state of the user to obtain the classification result, which includes the physiological state and psychological state of the user;

[0124] Selection unit 7023: used to process the classification results with an adaptive learning algorithm, adjust the magnetic field parameters, set the reward value according to the feedback of the user status, use the reinforcement learning algorithm to optimize the strategy, select the optimal action in each state to maximize the cumulative reward, repeatedly train in a simulated environment, update the Q value until convergence, and select the optimal magnetic field parameters in the current state according to the trained Q value function, wherein the user feedback includes the treatment effect and comfort.

[0125] Specifically, the processing module 703 includes:

[0126] The acquisition unit 7031 is used to acquire the user's real-time brain wave data, synchronize the magnetic field parameters and the user's real-time brain wave data to a unified time axis, and obtain combined data. The calculation formula is as follows:

[0127] D combined (t) = [EEG (t), B (t)]

[0128] Where, EEG(t) represents the brain wave data at time point t, and B(t) represents the magnetic field parameters at time point t;

[0129] Feature extraction unit 7032: used to model and extract features of the time series using an autoregressive model according to the combined data, obtain features of the brain dynamic trajectory, and obtain a feature vector through the coefficients and prediction residuals of the AR model;

[0130] Prediction unit 7033: used to predict the future brain dynamic trajectory according to the extracted feature vector through recursive neural network processing to obtain predicted brain wave data;

[0131] Acquisition unit 7034: Based on the predicted brain wave data, after reinforcement learning and deep Q network, the magnetic field parameters are continuously adjusted according to the reward mechanism to find the optimal strategy, and finally the optimal dynamic magnetic field parameters are obtained. The calculation formula is as follows:

[0132]

[0133] Where Q(s,a) is the value of taking action a in state s, α is the learning rate, r is the reward value, γ is the discount factor, s′ and a′ are the next state and action.

[0134] Specifically, the adjustment module 704 includes:

[0135] Input unit 7041: used to input the user's real-time brain wave data and the optimal dynamic magnetic field parameters into the fuzzy logic system, and fuzzify the input data through the membership function to obtain the fuzzified input data;

[0136] Fuzzy processing unit 7042: used to obtain fuzzy output data based on the fuzzified input data through fuzzy reasoning and fuzzy rule base processing, and obtain adjusted magnetic field parameters based on the fuzzy output data through defuzzification processing.

[0137] Specifically, the control module 705 includes:

[0138] Cluster processing unit 7051: used to process the personalized optimized magnetic field parameters and the acquired real-time brain wave data of the user by multi-dimensional data fusion technology, align the magnetic field parameters and brain wave data on a unified time axis, and fuse the data of different dimensions into a unified multi-dimensional data set by cluster analysis method, wherein the clustering process includes: selecting a suitable number of clusters k, clustering the data, randomly selecting the initial cluster center, iteratively updating the clusters and cluster centers to which the data points belong, until convergence, to obtain the fused multi-dimensional data set, and the calculation formula is as follows:

[0139]

[0140] In the formula, J is the clustering objective function, k is the number of clusters, x is the data point, and C i is the i-th cluster, u i is the center of the i-th cluster;

[0141] Feature extraction unit 7052: used to extract key features from the EEG data using wavelet transform method based on the fused data set to obtain EEG features, wherein the EEG signal is decomposed into different frequency bands, time-frequency features are extracted, and time series data containing feature information is obtained;

[0142] Construction unit 7053: used to use brain wave features as input data for dynamic time warping, align time series, calculate the minimum distance between two time series, and obtain the aligned time series; embed the data of the aligned time series into the high-dimensional phase space, select appropriate delay time and embedding dimension, convert the time series into a trajectory in the phase space, and use a deep learning algorithm to extract spatial features in the space and temporal features in the time series for fusion, so as to construct a dynamic graph, wherein the dynamic graph includes the connection relationship between different nodes and edges and the magnetic field parameters used to quantify each position of the user.

[0143] Specifically, the control module 705 further includes:

[0144] Establishment unit 7054: used to establish a three-dimensional finite element model of the magnetic therapy bed by using brain wave characteristics, pressure, temperature and magnetic field distribution data in the dynamic spectrum, and determine boundary conditions and load conditions, including electric field strength, frequency and physiological parameters of the user;

[0145] Solving unit 7055: used to solve the electric field distribution using the finite element method, and calculate the electric field strength and frequency distribution at each position; based on the electric field strength and frequency distribution, the ant colony optimization algorithm is used to obtain the optimal electric field strength and optimal frequency parameters, wherein the processing process includes each ant selecting a path according to the pheromone concentration and the heuristic function, constructing candidate solutions for the electric field strength and frequency, updating the pheromone concentration on the path according to the objective function value found by each ant, and repeating the path selection and pheromone update process until the stop condition is met;

[0146] Control unit 7056: used to input the optimal electric field strength and optimal frequency parameters into the real-time feedback system. The real-time feedback system monitors the user's physiological data, dynamically adjusts the magnetic field parameters based on the optimization results, and uses the fuzzy control algorithm for fine-tuning, thereby optimizing the control of the magnetic therapy bed.

[0147] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0148] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0149] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A brainwave-driven magnetic therapy bed optimization control method, characterized in that: include: Acquire the physiological data and environmental parameter data of the magnetic therapy bed user, including environmental status data, brain wave data, and electric field parameter data, and obtain a standardized data set after data preprocessing and normalization; Based on standardized data sets, biofeedback analysis is performed using machine learning algorithms and adaptive learning algorithms. After state analysis and parameter optimization, the specific parameters of the user's physiological and psychological state and the optimized magnetic field parameters are obtained. The magnetic field parameters are combined with the user's real-time brain wave data, and the dynamic trajectory of the brain is predicted using time series analysis and recursive neural networks. After being processed by an adaptive learning algorithm, the intensity and frequency of the magnetic field are dynamically adjusted to obtain the optimal dynamic magnetic field parameters. According to the optimal dynamic magnetic field parameters, the magnetic field parameters are further adjusted using the fuzzy control algorithm to obtain personalized optimized magnetic field parameters; Multi-dimensional data fusion technology is used to fuse the personalized optimized magnetic field parameters. A dynamic map is established through brain wave feature extraction and dynamic time warping algorithm. Combined with the data collected in real time by the electric field sensor, the optimal electric field strength and frequency are calculated through finite element analysis and ant colony optimization algorithm. Through the real-time feedback system, the magnetic field parameters are dynamically adjusted to achieve optimized control of the magnetic therapy bed.

2. The brainwave-driven magnetic therapy bed optimization control method according to claim 1 is characterized in that: Based on the standardized data set, the machine learning algorithm and the adaptive learning algorithm are used to perform biofeedback analysis, and after state analysis and parameter optimization, specific parameters of the user's physiological and psychological state and optimized magnetic field parameters are obtained, including: According to the environmental state data, brain wave data and electric field parameter data, the autoregressive model is used to extract the time series features to obtain the autoregressive coefficients, and each time series data is subjected to wavelet transformation to decompose it into different frequency components, and the wavelet coefficients are extracted as features. The time series data is subjected to Fourier transformation to obtain frequency domain features and form a high-dimensional feature vector; Based on the high-dimensional feature vector, a multi-layer perceptron neural network model is constructed, and the back-propagation algorithm and gradient descent optimizer are used to train the neural network, adjust the weights and biases, and after the training is completed, the new feature vector is input, and the neural network outputs the predicted user's physiological and psychological state to obtain the classification result, which includes the user's physiological and psychological state; The classification results are processed by an adaptive learning algorithm, the magnetic field parameters are adjusted, the reward value is set according to the user's status feedback, and the reinforcement learning algorithm is used to optimize the strategy. The optimal action is selected in each state to maximize the cumulative reward. The training is repeated in a simulated environment, and the Q value is updated until convergence. Based on the trained Q value function, the optimal magnetic field parameters in the current state are selected, and the user's feedback includes treatment effect and comfort.

3. The brainwave-driven magnetic therapy bed optimization control method according to claim 1 is characterized in that: The magnetic field parameters are combined with the user's real-time brain wave data, and the brain dynamic trajectory is predicted using time series analysis and recursive neural networks. After being processed by an adaptive learning algorithm, the intensity and frequency of the magnetic field are dynamically adjusted to obtain the optimal dynamic magnetic field parameters, including: The user's real-time brain wave data is obtained, and the magnetic field parameters and the user's real-time brain wave data are synchronized to a unified time axis to obtain combined data. The calculation formula is as follows: Dcombined(t)=[EEG(t),B(t)] Where, EEG(t) represents the brain wave data at time point t, and B(t) represents the magnetic field parameters at time point t; Based on the combined data, the autoregressive model is used to model and extract features of the time series to obtain the characteristics of the brain dynamic trajectory. The characteristic vector is obtained through the coefficients and prediction residuals of the AR model. According to the extracted feature vectors, the future brain dynamic trajectory is predicted through recursive neural network processing to obtain the predicted brain wave data; Based on the predicted brain wave data, after reinforcement learning and deep Q network, the magnetic field parameters are continuously adjusted according to the reward mechanism to find the optimal strategy, and finally the optimal dynamic magnetic field parameters are obtained. The calculation formula is as follows: Where Q(s,a) is the value of taking action a in state s, α is the learning rate, r is the reward value, γ is the discount factor, s′ and a′ are the next state and action.

4. The brainwave-driven magnetic therapy bed optimization control method according to claim 1, characterized in that: The method further adjusts the magnetic field parameters using a fuzzy control algorithm based on the optimal dynamic magnetic field parameters to obtain personalized optimized magnetic field parameters, including: The user's real-time brain wave data and optimal dynamic magnetic field parameters are input into the fuzzy logic system, and the input data is fuzzified through the membership function to obtain the fuzzified input data; According to the fuzzified input data, fuzzy output data is obtained through fuzzy reasoning and fuzzy rule base processing, and based on the fuzzy output data, the adjusted magnetic field parameters are obtained through defuzzification processing.

5. The brainwave-driven magnetic therapy bed optimization control method according to claim 1, characterized in that: The multi-dimensional data fusion technology is used to fuse the personalized optimized magnetic field parameters, and a dynamic map is established through brain wave feature extraction and dynamic time warping algorithm, including: The personalized optimized magnetic field parameters and the acquired real-time brain wave data of the user are processed by multi-dimensional data fusion technology, the magnetic field parameters and brain wave data are aligned to a unified time axis, and the data of different dimensions are fused into a unified multi-dimensional data set through clustering analysis method. The clustering process includes: selecting an appropriate number of clusters k, clustering the data, randomly selecting the initial cluster center, iteratively updating the clusters and cluster centers to which the data points belong until convergence, and obtaining the fused multi-dimensional data set. The calculation formula is as follows: In the formula, J is the clustering objective function, k is the number of clusters, x is the data point, and C i is the i-th cluster, u i is the center of the i-th cluster; Based on the fusion data set, the wavelet transform method is used to extract key features from the EEG data to obtain EEG features, wherein the EEG signal is decomposed into different frequency bands, the time-frequency features are extracted, and the time series data containing feature information is obtained; The brain wave characteristics are used as the input data of dynamic time warping. By aligning the time series, the minimum distance between the two time series is calculated to obtain the aligned time series. The data of the aligned time series is embedded in the high-dimensional phase space, and the appropriate delay time and embedding dimension are selected to transform the time series into a trajectory in the phase space. The deep learning algorithm is used to extract the spatial features in the space and the temporal features in the time series for fusion, so as to construct a dynamic graph, which includes the connection relationship between different nodes and edges and the magnetic field parameters used to quantify the user's position.

6. The brainwave-driven magnetic therapy bed optimization control method according to claim 1, characterized in that: The data collected in real time by the electric field sensor is combined with finite element analysis and ant colony optimization algorithm to calculate the optimal electric field strength and frequency. The magnetic field parameters are dynamically adjusted through a real-time feedback system to achieve optimal control of the magnetic therapy bed, including: Using brain wave characteristics, pressure, temperature and magnetic field distribution data in dynamic graphs, a three-dimensional finite element model of the magnetic therapy bed was established to determine boundary conditions and load conditions, including electric field strength, frequency and user's physiological parameters; The finite element method is used to solve the electric field distribution, and the electric field strength and frequency distribution at each position are calculated; based on the electric field strength and frequency distribution, the ant colony optimization algorithm is used to obtain the optimal electric field strength and optimal frequency parameters, wherein the processing process includes each ant selecting a path according to the pheromone concentration and the heuristic function, constructing candidate solutions for the electric field strength and frequency, updating the pheromone concentration on the path according to the objective function value found by each ant, and repeating the path selection and pheromone update process until the stop condition is met; The optimal electric field strength and optimal frequency parameters are input into the real-time feedback system. The real-time feedback system monitors the user's physiological data, dynamically adjusts the magnetic field parameters based on the optimization results, and uses a fuzzy control algorithm for fine-tuning, thereby optimizing the control of the magnetic therapy bed.

7. A brainwave-driven magnetic therapy bed optimization control system, based on the brainwave-driven magnetic therapy bed optimization control method according to claim 1, characterized in that: include: Acquisition module: used to obtain the physiological data and environmental parameter data of the magnetic therapy bed user, including environmental status data, brain wave data and electric field parameter data, and obtain a standardized data set after data preprocessing and normalization; Optimization module: used to perform biofeedback analysis based on standardized data sets, using machine learning algorithms and adaptive learning algorithms. After state analysis and parameter optimization, the specific parameters of the user's physiological and psychological state and the optimized magnetic field parameters are obtained; Processing module: used to combine magnetic field parameters with the user's real-time brain wave data, use time series analysis and recursive neural network to predict brain dynamic trajectory, and dynamically adjust the intensity and frequency of the magnetic field through adaptive learning algorithm processing to obtain the optimal dynamic magnetic field parameters; Adjustment module: used to further adjust the magnetic field parameters according to the optimal dynamic magnetic field parameters using fuzzy control algorithm to obtain personalized optimized magnetic field parameters; Control module: It is used to fuse the personalized optimized magnetic field parameters using multi-dimensional data fusion technology, establish a dynamic map through brain wave feature extraction and dynamic time warping algorithm, and combine the data collected in real time by the electric field sensor. Through finite element analysis and ant colony optimization algorithm, the optimal electric field strength and frequency are calculated, and the magnetic field parameters are dynamically adjusted through the real-time feedback system to achieve optimized control of the magnetic therapy bed.

8. The brainwave-driven magnetic therapy bed optimization control system according to claim 7 is characterized in that: The optimization module includes: Decomposition and transformation unit: used to extract time series features using the autoregressive model according to the environmental state data, brain wave data and electric field parameter data, obtain autoregressive coefficients, perform wavelet transformation on each time series data, decompose it into different frequency components, extract wavelet coefficients as features, perform Fourier transformation on the time series data, obtain frequency domain features, and form a high-dimensional feature vector; Training unit: used to build a multi-layer perceptron neural network model based on high-dimensional feature vectors, and use the back-propagation algorithm and gradient descent optimizer to train the neural network, adjust the weights and biases. After the training is completed, the new feature vector is input, and the neural network outputs the predicted user's physiological and psychological state to obtain the classification result, which includes the user's physiological and psychological state; Selection unit: used to process the classification results with an adaptive learning algorithm, adjust the magnetic field parameters, set the reward value based on the user's state feedback, use the reinforcement learning algorithm to optimize the strategy, select the optimal action in each state to maximize the cumulative reward, repeatedly train in a simulated environment, update the Q value until convergence, and select the optimal magnetic field parameters in the current state based on the trained Q value function, where the user's feedback includes treatment effect and comfort.

9. The brainwave-driven magnetic therapy bed optimization control system according to claim 7, characterized in that: The processing module includes: Acquisition unit: used to acquire the user's real-time brain wave data, synchronize the magnetic field parameters and the user's real-time brain wave data to a unified time axis, and obtain combined data. The calculation formula is as follows: Dcombined(t)=[EEG(t),B(t)] Where, EEG(t) represents the brain wave data at time point t, and B(t) represents the magnetic field parameters at time point t; Feature extraction unit: used to model and extract features of time series based on combined data using an autoregressive model, obtain the features of brain dynamic trajectory, and obtain feature vectors through the coefficients and prediction residuals of the AR model; Prediction unit: used to predict the future brain dynamic trajectory based on the extracted feature vectors through recursive neural network processing to obtain predicted brain wave data; Acquisition unit: Based on the predicted brain wave data, after reinforcement learning and deep Q network, the magnetic field parameters are continuously adjusted according to the reward mechanism to find the optimal strategy and finally obtain the optimal dynamic magnetic field parameters. The calculation formula is as follows: Where Q(s,a) is the value of taking action a in state s, α is the learning rate, r is the reward value, γ is the discount factor, s′ and a′ are the next state and action.

10. The brainwave-driven magnetic therapy bed optimization control system according to claim 7, characterized in that: The adjustment module includes: Input unit: used to input the user's real-time brain wave data and optimal dynamic magnetic field parameters into the fuzzy logic system, and fuzzify the input data through the membership function to obtain the fuzzified input data; Fuzzy processing unit: used to obtain fuzzy output data based on the fuzzified input data through fuzzy reasoning and fuzzy rule base processing, and to obtain adjusted magnetic field parameters based on the fuzzy output data through defuzzification processing.