Rehabilitation treatment system based on acupoint stimulation

Abstract

The invention relates to the technical field of acupoint stimulation, in particular to a rehabilitation treatment system based on acupoint stimulation, which comprises an acupoint network construction module, a community dynamic identification module, a treatment path optimization module, a stimulation parameter optimization module, an effect prediction module and a result feedback and adjustment module. According to the invention, through application of the shortest path algorithm, the most effective acupoint combination path for treating various symptoms can be accurately evaluated and determined, and through optimization of the treatment path, it is ensured that treatment measures can be concentrated on the acupoints with positive effects, the overall success rate of rehabilitation treatment is improved; the depth, frequency and duration of acupoint stimulation can be dynamically adjusted and optimized, stimulation parameters most suitable for specific patients are determined, the treatment effect is improved, specific requirements of different patients are met, treatment is more accurate and effective, and various complex and variable treatment requirements can be better met.

Description

Technical Field

[0001] The present invention relates to the technical field of acupoint stimulation, and in particular to a rehabilitation treatment system based on acupoint stimulation. Background Art

[0002] The field of acupoint stimulation technology involves the use of various methods to activate acupoints on the body. Acupoints are usually selected based on the meridian theory of traditional Chinese medicine. The techniques include acupuncture, electroacupuncture and pressure stimulation. The purpose is to regulate body functions, enhance self-healing ability, relieve pain, and improve health by stimulating specific acupoints. The technology integrates modern medical research with traditional treatment methods to develop a variety of devices and systems to provide standardized and replicable treatment effects.

[0003] Among them, the rehabilitation therapy system based on acupoint stimulation aims to provide an integrated solution to enhance the treatment effect and optimize the rehabilitation process. The system aims to improve the pertinence and efficiency of treatment by precisely stimulating specific acupoints, thereby accelerating the patient's recovery. The purpose is to achieve symptom relief, functional recovery and overall health improvement through a more systematic approach combined with modern technology to monitor and regulate the treatment process.

[0004] Traditional acupoint stimulation systems rely on standardized treatment plans and cannot achieve optimal treatment effects when treating patients with individual differences. For example, traditional methods use consistent stimulation parameters without fully considering the physiological differences and treatment responses between patients, resulting in poor treatment effects or prolonged treatment time. Traditional systems lack flexibility and adaptability in acupoint selection and stimulation parameter settings, making it difficult to quickly adjust treatment strategies for specific symptoms. In actual operations, this leads to low treatment efficiency and waste of resources, resulting in low patient satisfaction during treatment. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a rehabilitation treatment system based on acupoint stimulation.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a rehabilitation treatment system based on acupoint stimulation, the system comprising:

[0007] The acupoint network construction module is based on the patient's acupoint stimulation treatment records, defines the differentiated acupoints as nodes of the network, and defines the correlation between the treatment effects as the connection weights between the nodes to obtain the acupoint association structure;

[0008] The community dynamic identification module analyzes the connection nodes in the network based on the acupoint association structure, identifies acupoint groups with linkage effects, and obtains key acupoint effect groups;

[0009] The treatment path optimization module uses the shortest path algorithm based on the key acupoint effect groups to evaluate the therapeutic effects of differentiated acupoint combinations on target symptoms and obtain the recommended acupoint stimulation paths;

[0010] The stimulation parameter optimization module iteratively updates the acupoint stimulation parameters based on the acupoint stimulation suggested path through a genetic algorithm to obtain an improved acupoint stimulation strategy;

[0011] The effect prediction module predicts and analyzes the patient's target acupoint stimulation response based on the improved acupoint stimulation strategy to obtain an acupoint stimulation prediction result;

[0012] The result feedback and adjustment module collects the real-time feedback data of the patient during the acupoint stimulation rehabilitation treatment based on the acupoint stimulation prediction results and the improved acupoint stimulation strategy, optimizes the acupoint stimulation to match the patient's actual response, and obtains treatment suggestions for adjustment and optimization.

[0013] The present invention has improvements in that the acupoint association structure includes quantitative attributes of nodes, treatment association weights between nodes and network connectivity indicators, the key acupoint effect group includes key active acupoint groups, similar treatment response acupoint combinations and interaction strengths of acupoint groups in treatment, the acupoint stimulation suggestion path includes acupoint stimulation sequence, expected treatment effect improvement ratio and time optimization during treatment, the improved acupoint stimulation strategy includes adjusted acupoint stimulation depth, frequency and duration, the acupoint stimulation prediction results include predicted treatment effects, reaction sensitivities of key acupoints and predictions of effect changes under patient target treatment conditions, and the adjusted and optimized treatment suggestions include acupoint selection adjusted according to real-time data, stimulation parameter fine-tuning and improvement indicators of expected rehabilitation treatment effects.

[0014] The present invention is improved in that the acupoint network construction module comprises:

[0015] The data integration submodule collects and classifies the acupoint stimulation intensity, duration and frequency data of each treatment of the patient based on the patient's acupoint stimulation treatment records, performs data cleaning and format unification, optimizes the data quality and consistency, and obtains the acupoint treatment data set;

[0016] The graph theory construction submodule defines the acupoints in each independent treatment record as nodes of the network based on the acupoint treatment data set, defines the connections between the nodes based on the fit and difference of the treatment cases, constructs the connections between the nodes, and forms an acupoint network structure diagram;

[0017] The network weight configuration submodule analyzes the interactions between nodes based on the acupoint network structure diagram, calculates the connection weights according to the correlation of the rehabilitation treatment effects of acupoint stimulation, adjusts the strength of the network connections, reflects the influence of differentiated acupoints on the treatment effects, optimizes and improves the network structure, and obtains the acupoint association structure.

[0018] The present invention is improved in that the community dynamic identification module includes:

[0019] The graph analysis submodule analyzes the network structure based on the acupoint association structure, identifies the key nodes in the network and the connection relationship between the key nodes, calculates the centrality and density structure characteristics of the network, and obtains the network structure characteristic analysis results;

[0020] The community detection submodule groups the nodes in the network based on the results of the network structure feature analysis, identifies acupoint groups with similar therapeutic effects and therapeutic responses, analyzes the strength of association between acupoint groups, identifies acupoint groups with therapeutic linkage effects, and obtains effect group identification results;

[0021] The association effect evaluation submodule quantitatively evaluates the therapeutic linkage effect between acupoint groups based on the effect group identification results, compares the therapeutic effect differences between differentiated acupoint groups, identifies key therapeutic response acupoint combinations, and obtains key acupoint effect groups.

[0022] The present invention is improved in that the treatment path optimization module comprises:

[0023] The pathway analysis submodule analyzes the correlation between acupoints based on the key acupoint effect groups, uses the shortest path algorithm to evaluate the therapeutic effects of differentiated acupoint combinations on target symptoms, identifies key therapeutic pathways by analyzing the historical treatment success rate of each acupoint combination and the interaction strength between acupoints, and generates therapeutic pathway mapping information;

[0024] The efficacy impact assessment submodule evaluates the efficacy mobility of each treatment pathway based on the treatment pathway mapping information, calculates the contribution of each pathway to the overall treatment effect, quantifies and compares the efficacy contributions of multiple pathways, identifies the treatment pathways with key influence, and obtains the efficacy contribution analysis results;

[0025] The path selection submodule evaluates the actual feasibility and expected therapeutic effect of the path based on the efficacy contribution analysis result, selects an acupoint stimulation path that matches the expected therapeutic efficiency and efficacy contribution, and obtains a recommended acupoint stimulation path.

[0026] The present invention is improved in that the shortest path algorithm is according to the formula:

[0027] d[v]=d[u]+αw(u,v)+βe(u,v)+γs(u,v)

[0028] The shortest path distance between acupoints was calculated, where d[v] was the shortest path distance from the source acupoint to acupoint v, d[u] was the known shortest path distance from the source acupoint to acupoint u, w(u,v) represented the basic treatment effect weight between node u and node v, e(u,v) represented the complementary treatment effect enhancement factor between acupoints u and v, s(u,v) represented the patient satisfaction influencing factor, α was the adjustment coefficient of the basic treatment effect weight, β was the adjustment coefficient of the complementary treatment effect enhancement factor, and γ was the adjustment coefficient of the patient satisfaction influencing factor.

[0029] The present invention is improved in that the stimulation parameter optimization module comprises:

[0030] The parameter simulation submodule performs acupoint stimulation parameter simulation based on the acupoint stimulation suggested path, wherein the parameter simulation includes multiple stimulation depths, frequencies and durations, evaluates the potential impact of differentiated parameter combinations on the treatment effect, and obtains acupoint stimulation simulation results;

[0031] The optimization and adjustment submodule optimizes the acupoint stimulation parameters based on the acupoint stimulation simulation results by using a genetic algorithm, captures the parameter configuration of the target treatment effect by iteratively optimizing the parameter combination, and obtains the optimized parameter configuration result;

[0032] The parameter setting submodule sets and adjusts the acupoint stimulation parameters based on the optimization parameter configuration result to match the expected treatment effect, optimize the efficiency and results of the rehabilitation treatment process, and obtain an improved acupoint stimulation strategy.

[0033] The present invention is improved in that the genetic algorithm is according to the formula:

[0034]

[0035] Calculate the acupoint stimulation parameters, where is the parameter value of individual i in the next generation, is the parameter value of individual i in the current generation, is the parameter value of the best performing individual in the current generation, is the parameter value of the first randomly selected individual in the current generation, is the parameter value of the second individual randomly selected in the current generation, F is the scaling factor, and λ is the adjustment factor.

[0036] Compared with the prior art, the advantages and positive effects of the present invention are:

[0037] In the present invention, through the application of the shortest path algorithm, the most effective acupoint combination path for treating various symptoms can be accurately evaluated and determined. By optimizing the treatment path, it is ensured that the treatment measures can be concentrated on the acupoints that produce positive effects, thereby improving the overall success rate of rehabilitation treatment. The use of genetic algorithms can dynamically adjust and optimize the depth, frequency and duration of acupoint stimulation, determine the stimulation parameters that are most suitable for specific patients, improve the treatment effect, adapt to the specific needs of different patients, be more precise and effective in providing treatment, and better adapt to various complex and changeable treatment needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A module diagram of a rehabilitation treatment system based on acupoint stimulation is proposed for the present invention;

[0039] Figure 2 The present invention proposes a system framework diagram of a rehabilitation treatment system based on acupoint stimulation;

[0040] Figure 3 A schematic diagram of an acupoint network construction module in a rehabilitation treatment system based on acupoint stimulation proposed by the present invention;

[0041] Figure 4 A schematic diagram of a community dynamic identification module in a rehabilitation treatment system based on acupoint stimulation proposed by the present invention;

[0042] Figure 5 A schematic diagram of a treatment path optimization module in a rehabilitation treatment system based on acupoint stimulation proposed by the present invention;

[0043] Figure 6 A schematic diagram of a stimulation parameter optimization module in a rehabilitation treatment system based on acupoint stimulation proposed by the present invention;

[0044] Figure 7 A schematic diagram of an effect prediction module in a rehabilitation treatment system based on acupoint stimulation proposed by the present invention;

[0045] Figure 8 The present invention provides a schematic diagram of a result feedback and adjustment module in a rehabilitation treatment system based on acupoint stimulation. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0048] Example

[0049] See also Figure 1 , the present invention provides a technical solution: a rehabilitation treatment system based on acupoint stimulation, the system includes an acupoint network construction module, a community dynamic identification module, a treatment path optimization module, a stimulation parameter optimization module, an effect prediction module, and a result feedback and adjustment module;

[0050] The acupoint network construction module is based on the patient's acupoint stimulation treatment record, which includes the acupoint stimulation intensity, duration and stimulation frequency data in the treatment case. Differentiated acupoints are defined as nodes of the network, and the correlation between treatment effects is defined as the connection weight between nodes. A weighted undirected graph that represents the mutual influence between acupoints is constructed to obtain the acupoint association structure.

[0051] The community dynamic identification module analyzes the connection nodes in the network based on the acupoint association structure, identifies acupoint groups with linkage effects, and obtains key acupoint effect groups;

[0052] The treatment path optimization module uses the shortest path algorithm based on the key acupoint effect groups to evaluate the therapeutic effects of differentiated acupoint combinations on target symptoms, and identifies effective acupoint stimulation delivery paths through network flow analysis to obtain recommended acupoint stimulation paths;

[0053] The stimulation parameter optimization module optimizes the depth, frequency, and duration parameters of acupoint stimulation based on the acupoint stimulation suggested path through genetic algorithms, iteratively updates the acupoint stimulation parameters, captures the target parameter configuration, and obtains an improved acupoint stimulation strategy;

[0054] The effect prediction module is based on the improved acupoint stimulation strategy, combined with the patient's health data and historical parameters of acupoint stimulation, to predict and analyze the patient's target acupoint stimulation response and obtain the acupoint stimulation prediction results;

[0055] The result feedback and adjustment module collects real-time feedback data from patients during acupoint stimulation rehabilitation treatment based on the acupoint stimulation prediction results and improved acupoint stimulation strategies, dynamically adjusts the treatment strategies, optimizes acupoint stimulation to match the patients' actual responses, and obtains adjusted and optimized treatment recommendations.

[0056] The acupoint association structure includes the quantitative attributes of nodes, the therapeutic association weights between nodes and the network connectivity indicators. The key acupoint effect groups include the key active acupoint groups, the combination of similar therapeutic response acupoints and the interaction intensity of acupoint groups in treatment. The recommended acupoint stimulation paths include the acupoint stimulation sequence, the expected improvement ratio of therapeutic effect and the time optimization during the treatment process. The improved acupoint stimulation strategies include the adjusted acupoint stimulation depth, frequency and duration. The acupoint stimulation prediction results include the predicted therapeutic effect, the response sensitivity of key acupoints and the prediction of the effect change under the patient's target treatment conditions. The adjusted and optimized treatment recommendations include the acupoint selection adjusted according to real-time data, the fine-tuning of stimulation parameters and the improvement indicators of the expected rehabilitation treatment effect.

[0057] See also Figure 2 and Figure 3 ,The acupoint network construction module includes a data integration submodule, a graph theory construction submodule, and a network weight configuration submodule;

[0058] The data integration submodule collects and classifies the acupoint stimulation intensity, duration and frequency data of each treatment of the patient based on the patient's acupoint stimulation treatment records, performs data cleaning and format unification, optimizes the quality and consistency of the data, and obtains the specific process of the acupoint treatment data set as follows;

[0059] The data integration submodule is based on the patient's acupoint stimulation treatment records and adopts data preprocessing technology. It performs data cleaning through Python's Pandas library, uses the dropna() function to remove rows containing missing values ​​to ensure data integrity, and uses the astype() function to convert all numerical data into a unified format. For example, all time records are unified into minutes. The query() function is used to exclude abnormal data points that are beyond the normal treatment range, such as records with abnormally high stimulation intensity, to improve the quality and consistency of the data, lay the foundation for subsequent analysis, and generate an acupoint treatment data set.

[0060] The graph theory construction submodule is based on the acupoint treatment data set. The acupoints in each independent treatment record are defined as nodes of the network. The connections between nodes are defined based on the fit and difference of the treatment cases. The connections between nodes are constructed to form the specific process of the acupoint network structure diagram.

[0061] The graph theory construction submodule is based on the acupoint treatment dataset and applies graph theory algorithms to construct an acupoint network model. Through Python's NetworkX library, the add_node() function is used to add each independent acupoint data as a node of the network. According to the similarity and difference of treatment cases between acupoints, the add_edge() function is used to establish connections between nodes. The existence of connections depends on the fit evaluation of the treatment cases, which is determined by calculating the co-occurrence frequency and treatment effect differences of different acupoint combinations. A graph containing each acupoint as a node is formed, and the potential treatment relationship between acupoints is reflected through connections to generate an acupoint network structure diagram.

[0062] The network weight configuration submodule is based on the acupoint network structure diagram, analyzes the interaction between nodes, calculates the connection weight according to the correlation of the rehabilitation treatment effect of acupoint stimulation, adjusts the strength of the network connection, reflects the influence of differentiated acupoints on the treatment effect, optimizes and improves the network structure, and obtains the specific process of the acupoint association structure as follows;

[0063] The network weight configuration submodule performs network weight configuration based on the acupoint network structure diagram, optimizes the network diagram by adjusting the weights of the connections between nodes, and uses the set_edge_attributes() function of the Python and NetworkX libraries to assign weights to each connection. The weights are calculated based on the correlation between the rehabilitation treatment effects of acupoint stimulation. The calculation method includes considering the patient's recovery speed and the durability of the treatment effect after acupoint stimulation, and using the force-directed graph layout algorithm spring_layout() to optimize the layout of nodes and edges, so that the node positions and edge thicknesses in the graph intuitively express the therapeutic influence between acupoints and generate an acupoint association structure.

[0064] See also Figure 2 and Figure 4 ,The community dynamic identification module includes a graph analysis submodule, a community detection submodule, and an association effect evaluation submodule;

[0065] The graph analysis submodule analyzes the network structure based on the acupoint association structure, identifies the key nodes in the network and the connection relationship between the key nodes, calculates the centrality and density structure characteristics of the network, and obtains the specific process of the network structure characteristic analysis results as follows;

[0066] The graph analysis submodule uses network analysis technology based on the acupoint association structure and Python's NetworkX library to perform graph analysis and calculate the centrality of each node in the network.

[0067] The betweenness_centrality() function evaluates the importance of acupuncture point nodes in the entire network, and uses the density() function to calculate the density of the network. The density index reflects the ratio of actual connections to estimated connections in the network, identifies key nodes and their connection relationships, and generates network structure feature analysis results.

[0068] The community detection submodule groups the nodes in the network based on the results of the network structure feature analysis, identifies acupoint groups with similar therapeutic effects and responses, analyzes the strength of association between acupoint groups, identifies acupoint groups with therapeutic linkage effects, and obtains the specific process of effect group identification results as follows;

[0069] The community detection submodule is based on the results of network structure feature analysis, adopts community detection algorithm, uses Python's community library to implement Louvain method, mines the community structure in the network by optimizing modularity, calls the best_partition() function, divides the nodes in the network into multiple communities, identifies acupoint groups with similar therapeutic effects and responses, and calculates the connection density within and between communities, analyzes the association strength between groups, identifies acupoint groups with significant therapeutic linkage effects, and generates effect group identification results.

[0070] The association effect evaluation submodule quantitatively evaluates the therapeutic linkage effect between acupoint groups based on the effect group identification results, compares the therapeutic effect differences between differentiated acupoint groups, identifies key therapeutic response acupoint combinations, and obtains the specific process of key acupoint effect groups as follows;

[0071] The association effect evaluation submodule performs a quantitative evaluation of the therapeutic linkage effect between acupoint groups based on the effect group identification results. It uses statistical analysis methods and the stats module in Python's SciPy library to perform inter-group comparisons, including the use of the mannwhitneyu() function for non-parametric U tests, to compare the significant differences in therapeutic effects between different acupoint groups, identify acupoint combinations with significant therapeutic effects, and record the statistical scores of each acupoint combination in detail to help determine the most critical therapeutic response acupoint combination and generate key acupoint effect groups.

[0072] See also Figure 2 and Figure 5 ,The treatment pathway optimization module includes the pathway analysis submodule, the efficacy impact evaluation submodule, and the pathway selection submodule;

[0073] The pathway analysis submodule analyzes the correlation between acupoints based on the key acupoint effect groups, and uses the shortest path algorithm to evaluate the therapeutic effects of differentiated acupoint combinations on target symptoms. By analyzing the historical treatment success rate of each acupoint combination and the interaction strength between acupoints, the key treatment pathways are identified and the specific process of generating treatment pathway mapping information is as follows;

[0074] The path analysis submodule applies the shortest path algorithm based on the key acupoint effect groups. Through Python's NetworkX library, it calls the all_pairs_dijkstra_path() function to calculate the shortest paths between all possible acupoint combinations in the network. The function runs based on the Dijkstra algorithm and determines the path length by considering the edge weights between nodes. It helps evaluate the therapeutic effects of different acupoint combinations on target symptoms, identifies key treatment pathways through comprehensive path data, and generates treatment path mapping information.

[0075] The shortest path algorithm is based on the formula:

[0076] d[v]=d[u]+αw(u,v)+βe(u,v)+γs(u,v)

[0077] The shortest path distance between acupoints was calculated, where d[v] is the shortest path distance from the source acupoint to acupoint v, d[u] is the known shortest path distance from the source acupoint to acupoint u, w(u,v) represents the basic therapeutic effect weight between node u and node v, and the weight reflects the direct therapeutic connection strength between the two acupoints. e(u,v) represents the complementary therapeutic effect enhancement factor between acupoints u and v, and the factor measures the synergistic effect of the two acupoints when used in combination, considering whether the interaction between the two acupoints in treatment can enhance the overall effect of the treatment. s(u,v) represents the patient satisfaction influencing factor, which measures the patient's subjective feelings and objective treatment responses after stimulation of the u and v acupoints, and reflects the patient's acceptance and satisfaction with a specific treatment combination. α is the adjustment coefficient of the basic treatment effect weight, which is used to adjust the influence of w(u,v) in the overall pathway calculation. β is the adjustment coefficient of the complementary treatment effect enhancement factor, so that e(u,v) is properly considered in the pathway calculation. γ is the adjustment coefficient of the patient satisfaction influencing factor, which adjusts the weight of s(u,v) in the overall pathway evaluation.

[0078] The specific implementation process of the improved formula is as follows:

[0079] Initialize d[v] of all nodes v to infinity, set d[u] of the source node to 0, and use the all_pairs_dijkstra_path() method of the NetworkX library to apply the formula to iteratively calculate the shortest path between each node pair (u, v). Determine the basic weight w(u, v) through the analysis of treatment cases and patient feedback data. The complementary treatment effect enhancement factor e(u, v) is evaluated by the complementarity of the acupoint combination in clinical treatment cases. The patient satisfaction influencing factor s(u, v) is extracted from the patient's real-time feedback and post-treatment survey. The weight coefficients α, β and γ are determined by statistical analysis of historical treatment data to ensure that each factor is weighted according to its actual influence. In each iteration, select the smallest d[v] as the source node u for the next step. Repeat this process until the shortest paths of all nodes are found, obtain the predicted impact of each acupoint combination on the treatment of the target symptom, and generate treatment path mapping information.

[0080] The efficacy impact assessment submodule evaluates the efficacy mobility of each treatment pathway based on the treatment pathway mapping information. By calculating the contribution of each pathway to the overall treatment effect, the efficacy contribution of multiple pathways is quantified and compared, and the treatment pathways with key influence are identified. The specific process for obtaining the efficacy contribution analysis results is as follows;

[0081] The efficacy impact assessment submodule performs path efficacy fluidity assessment based on the treatment pathway mapping information, uses Python's NumPy library for numerical calculations, and uses the np.sum() and np.mean() functions to calculate the treatment contribution of each pathway. By analyzing the performance of each pathway in historical data, such as the statistical mean of the treatment success rate, the efficacy contribution of multiple pathways is quantified and compared, thereby identifying treatment pathways with key influence and obtaining efficacy contribution analysis results.

[0082] The path selection submodule evaluates the actual feasibility and expected treatment effect of the path based on the results of the efficacy contribution analysis, selects the acupoint stimulation path that matches the expected treatment efficiency and efficacy contribution, and obtains the specific process of the recommended acupoint stimulation path as follows;

[0083] The path selection submodule evaluates the actual feasibility and expected treatment effect of each path based on the results of efficacy contribution analysis. It uses decision analysis methods and Python's Pandas library to sort and filter data. It calls the DataFrame.sort_values() function to sort the treatment paths by efficacy contribution and selects the top-ranked paths. The paths take into account the optimal match between treatment efficiency and efficacy to obtain the recommended paths for acupoint stimulation.

[0084] See also Figure 2 and Figure 6,The stimulation parameter optimization module includes a parameter simulation submodule, an optimization adjustment submodule, and a parameter setting submodule;

[0085] The parameter simulation submodule simulates acupoint stimulation parameters based on the acupoint stimulation suggested path. The parameter simulation includes various stimulation depths, frequencies and durations, and evaluates the potential impact of differentiated parameter combinations on the treatment effect. The specific process of obtaining the acupoint stimulation simulation results is as follows;

[0086] The parameter simulation submodule performs simulation of acupoint stimulation parameters based on the suggested acupoint stimulation path. It uses simulation tools in MATLAB or Python to set up the simulation environment, define multiple combinations of acupoint stimulation depth, frequency, and duration, use loop statements to traverse the parameter combinations, and calculate the potential impact of each combination on the treatment effect through the simulation model. The evaluation of each parameter combination is based on a pre-set simulation algorithm to ensure that the efficacy prediction of each combination is calculated based on historical data and the treatment response model, thereby generating acupoint stimulation simulation results.

[0087] The optimization and adjustment submodule optimizes the acupoint stimulation parameters based on the acupoint stimulation simulation results through genetic algorithms, and captures the parameter configuration of the target treatment effect through iterative optimization of parameter combinations. The specific process of obtaining the optimized parameter configuration results is as follows;

[0088] The optimization and adjustment submodule is based on the simulation results of acupoint stimulation, and performs parameter optimization through genetic algorithms. It is implemented in the Python environment using the DEAP library, initializes the genetic population, including multiple combinations of stimulation depth, frequency, and duration, and then performs selection, crossover, and mutation operations. The therapeutic effect of each generation of population is evaluated through the fitness function, and multiple generations are iterated until the optimal parameter configuration is found. The parameter configuration updated in each iteration is based on the optimal result of the previous generation and random mutation, ensuring that the final parameter configuration is the most effective for the target therapeutic effect, and generating the optimized parameter configuration result.

[0089] Genetic algorithm, according to the formula:

[0090]

[0091] Calculate the acupoint stimulation parameters, where is the parameter value of individual i in the next generation, is the parameter value of individual i in the current generation, is the parameter value of the best performing individual in the current generation, which is used to guide the population to evolve towards a better solution. is the parameter value of the first randomly selected individual in the current generation, used to increase the diversity of the genetic algorithm. is the parameter value of the second individual randomly selected in the current generation, which is also used to increase diversity. F is the scaling factor, which controls the amplitude of mutation in the mutation operation. λ is the adjustment factor, which is used to adjust the influence of random mutation and enhance the exploration ability and adaptability of the algorithm.

[0092] The specific implementation process of the improved formula is as follows:

[0093] By initializing the genetic population, covering multiple combinations of stimulation depth, frequency and duration, selection, crossover and mutation operations are performed. The adjustment factor λ is introduced in the mutation operation. The adjustment factor λ is dynamically adjusted according to the evaluation of the treatment effect of the previous generations to ensure that the algorithm not only relies on the historical optimal solution, but also increases randomness to prevent premature convergence. The adjustment factor λ is used to adjust the degree of random mutation. Combined with the scaling factor F, according to Calculate the mutation value of each individual, use the fitness function to evaluate the treatment effect of each individual on the new generation of population, select the optimal individual based on the evaluation results, and update the population until the preset number of iterations is reached or a satisfactory optimal parameter configuration is found.

[0094] The parameter setting submodule sets and adjusts the acupoint stimulation parameters based on the optimized parameter configuration results to match the expected treatment effect, optimize the efficiency and results of the rehabilitation treatment process, and obtain the specific process of the improved acupoint stimulation strategy as follows;

[0095] The parameter setting submodule determines and adjusts the final parameters of acupoint stimulation based on the optimization parameter configuration results. It uses Python's Pandas library for data processing and adjusts the parameter values ​​according to the optimization results, such as adjusting the values ​​of the corresponding columns in the DataFrame to match the optimized stimulation depth, frequency, and duration. The parameters are set in the actual treatment equipment to ensure that each treatment is performed according to the optimized parameters, thereby improving the efficiency and effectiveness of the treatment process and generating an improved acupoint stimulation strategy.

[0096] See also Figure 2 and Figure 7 ,The effect prediction module includes a feature extraction submodule, a model training submodule, and an effect evaluation submodule;

[0097] The feature extraction submodule is based on the improved acupoint stimulation strategy, combined with the patient's health data and the historical parameters of acupoint stimulation, to perform data extraction and feature extraction, including the location of the acupoints, stimulation frequency, treatment history and the patient's basic health status, and screen the key features that have an impact on the prediction of acupoint stimulation response. The specific process of obtaining the feature data set is as follows;

[0098] The feature extraction submodule is based on the improved acupoint stimulation strategy, combined with the patient's health data and the historical parameters of acupoint stimulation, and uses Python's Pandas and Scikit-learn libraries for data processing and feature extraction. The DataFrame operation is used to merge patient information and treatment data, and the SelectKBest method is applied with the f_classif function to filter out the features most relevant to the prediction of acupoint stimulation response, such as acupoint location, stimulation frequency, treatment history, and basic health status of the patient. The importance of each feature is confirmed through statistical tests to ensure that the final feature set contains the most useful information for the prediction model and generate a feature data set.

[0099] The model training submodule builds a prediction model based on the feature data set, trains the prediction model, adjusts the model parameters, and performs cross-validation to optimize the model performance. The specific process of obtaining the prediction optimization model is as follows;

[0100] The model training submodule builds a prediction model based on the feature data set, and uses the random forest algorithm in the Scikit-learn library to build the model. The model is initialized through the RandomForestClassifier class, and parameters such as the number and depth of trees are set. The data is split using train_test_split to isolate the validation set. Cross_val_score is used for cross-validation, and the model parameters are adjusted through repeated training and validation cycles to optimize the model's prediction performance for unseen data, thereby ensuring that the model's generalization ability is maximized and a prediction optimization model is obtained.

[0101] The effect evaluation submodule predicts and analyzes the patient's target acupoint stimulation response based on the prediction optimization model and the improved acupoint stimulation strategy, predicts the rehabilitation treatment effect of the target acupoint stimulation, and obtains the specific process of the acupoint stimulation prediction result as follows;

[0102] The effect evaluation submodule performs the final effect evaluation of the model based on the prediction optimization model and the improved acupoint stimulation strategy. The predict() function is used to predict and analyze the stimulation response of the target acupoints. The accuracy and other performance indicators of the model are calculated through functions such as accuracy_score and confusion_matrix to analyze the accuracy and practicality of the prediction results. This helps to understand the potential effect of the acupoint stimulation strategy in actual treatment and obtain the prediction results of acupoint stimulation.

[0103] See also Figure 2 and Figure 8 ,The result feedback and adjustment module includes a feedback collection submodule, a strategy adjustment submodule, and a solution updating submodule;

[0104] The feedback collection submodule collects the patient's real-time feedback data during the acupoint stimulation rehabilitation treatment, including the patient's subjective feelings and objective physiological changes, based on the acupoint stimulation prediction results and the improved acupoint stimulation strategy, combined with the patient's health data and historical parameters of acupoint stimulation, and integrates and processes the data to obtain the specific process of the feedback data set:

[0105] The feedback collection submodule collects and integrates real-time feedback data through Python's Pandas library based on the acupoint stimulation prediction results and improved acupoint stimulation strategies, combined with the patient's health data and historical parameters of acupoint stimulation. The read_csv() function is used to import data from multiple source files, and the merge() function is used to merge the patient's treatment history and real-time health feedback data. Then, drop_duplicates() is used to remove duplicate records to ensure the accuracy and completeness of the data. The data including the patient's subjective feelings and objective physiological changes are comprehensively sorted to generate a feedback data set.

[0106] The strategy adjustment submodule analyzes the consistency between patient feedback and predicted results based on the feedback data set, adjusts acupoint stimulation parameters and treatment pathways, matches the patient's actual response, optimizes the effectiveness of acupoint stimulation rehabilitation therapy, and generates treatment strategy optimization suggestions. The specific process is as follows;

[0107] The strategy adjustment submodule uses the logistic regression model in Python's Scikit-learn library to analyze the consistency of feedback and prediction results based on the feedback data set. The model is initialized by calling LogisticRegression(), and solver = 'liblinear' is set to handle the optimization problem of small data sets. The model is trained using the fit() method, and the acupoint stimulation parameters and treatment paths are adjusted to match the actual patient response. The adjusted parameters are verified by the predict() function to ensure the highest matching degree and generate treatment strategy optimization suggestions.

[0108] The scheme update submodule updates the treatment suggestions based on the treatment strategy optimization suggestions, applies the adjusted acupoint stimulation strategy to the actual treatment, monitors the actual effect of the updated acupoint stimulation strategy, and continuously optimizes it. The specific process of obtaining the adjusted and optimized treatment suggestions is as follows;

[0109] The program update submodule updates treatment recommendations based on treatment strategy optimization suggestions and applies them to actual treatment. The treatment parameters are updated through Python's Pandas library, and the optimized strategy is saved as a new configuration file using the to_csv() function to ensure that all relevant equipment and treatment protocols are updated according to the latest parameters. The new strategy is applied to the treatment process in real time using the apply() function, and the actual effect after the update is monitored and recorded. Through continuous data collection and analysis, the treatment effect is optimized and the adjusted and optimized treatment recommendations are generated.

[0110] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A rehabilitation treatment system based on acupoint stimulation, characterized in that: The system comprises: The acupoint network construction module is based on the patient's acupoint stimulation treatment records, defines the differentiated acupoints as nodes of the network, and defines the correlation between the treatment effects as the connection weights between the nodes to obtain the acupoint association structure; The community dynamic identification module analyzes the connection nodes in the network based on the acupoint association structure, identifies acupoint groups with linkage effects, and obtains key acupoint effect groups; The treatment path optimization module uses the shortest path algorithm based on the key acupoint effect groups to evaluate the therapeutic effects of differentiated acupoint combinations on target symptoms and obtain the recommended acupoint stimulation paths; The stimulation parameter optimization module iteratively updates the acupoint stimulation parameters based on the acupoint stimulation suggested path through a genetic algorithm to obtain an improved acupoint stimulation strategy; The effect prediction module predicts and analyzes the patient's target acupoint stimulation response based on the improved acupoint stimulation strategy to obtain an acupoint stimulation prediction result; The result feedback and adjustment module collects the real-time feedback data of the patient during the acupoint stimulation rehabilitation treatment based on the acupoint stimulation prediction results and the improved acupoint stimulation strategy, optimizes the acupoint stimulation to match the patient's actual response, and obtains treatment suggestions for adjustment and optimization.

2. The rehabilitation treatment system based on acupoint stimulation according to claim 1, characterized in that: The acupoint association structure includes quantitative attributes of nodes, treatment association weights between nodes and network connectivity indicators; the key acupoint effect group includes key active acupoint groups, similar treatment response acupoint combinations and the interaction intensity of acupoint groups in treatment; the acupoint stimulation recommendation path includes acupoint stimulation sequence, expected treatment effect improvement ratio and time optimization during treatment; the improved acupoint stimulation strategy includes adjusted acupoint stimulation depth, frequency and duration; the acupoint stimulation prediction results include predicted treatment effects, response sensitivity of key acupoints and prediction of effect changes under patient target treatment conditions; the adjusted and optimized treatment recommendations include acupoint selection adjusted according to real-time data, stimulation parameter fine-tuning and improvement indicators of expected rehabilitation treatment effects.

3. The rehabilitation treatment system based on acupoint stimulation according to claim 1, characterized in that: The acupoint network building module includes: The data integration submodule collects and classifies the acupoint stimulation intensity, duration and frequency data of each treatment of the patient based on the patient's acupoint stimulation treatment records, performs data cleaning and format unification, optimizes the data quality and consistency, and obtains the acupoint treatment data set; The graph theory construction submodule defines the acupoints in each independent treatment record as nodes of the network based on the acupoint treatment data set, defines the connections between the nodes based on the fit and difference of the treatment cases, constructs the connections between the nodes, and forms an acupoint network structure diagram; The network weight configuration submodule analyzes the interactions between nodes based on the acupoint network structure diagram, calculates the connection weights according to the correlation of the rehabilitation treatment effects of acupoint stimulation, adjusts the strength of the network connections, reflects the influence of differentiated acupoints on the treatment effects, optimizes and improves the network structure, and obtains the acupoint association structure.

4. The rehabilitation treatment system based on acupoint stimulation according to claim 1, characterized in that: The community dynamics identification module includes: The graph analysis submodule analyzes the network structure based on the acupoint association structure, identifies the key nodes in the network and the connection relationship between the key nodes, calculates the centrality and density structure characteristics of the network, and obtains the network structure characteristic analysis results; The community detection submodule groups the nodes in the network based on the results of the network structure feature analysis, identifies acupoint groups with similar therapeutic effects and therapeutic responses, analyzes the strength of association between acupoint groups, identifies acupoint groups with therapeutic linkage effects, and obtains effect group identification results; The association effect evaluation submodule quantitatively evaluates the therapeutic linkage effect between acupoint groups based on the effect group identification results, compares the therapeutic effect differences between differentiated acupoint groups, identifies key therapeutic response acupoint combinations, and obtains key acupoint effect groups.

5. The rehabilitation treatment system based on acupoint stimulation according to claim 1, characterized in that: The treatment pathway optimization module includes: The pathway analysis submodule analyzes the correlation between acupoints based on the key acupoint effect groups, uses the shortest path algorithm to evaluate the therapeutic effects of differentiated acupoint combinations on target symptoms, identifies key therapeutic pathways by analyzing the historical treatment success rate of each acupoint combination and the interaction strength between acupoints, and generates therapeutic pathway mapping information; The efficacy impact assessment submodule evaluates the efficacy mobility of each treatment pathway based on the treatment pathway mapping information, calculates the contribution of each pathway to the overall treatment effect, quantifies and compares the efficacy contributions of multiple pathways, identifies the treatment pathways with key influence, and obtains the efficacy contribution analysis results; The path selection submodule evaluates the actual feasibility and expected therapeutic effect of the path based on the efficacy contribution analysis result, selects an acupoint stimulation path that matches the expected therapeutic efficiency and efficacy contribution, and obtains a recommended acupoint stimulation path.

6. The rehabilitation treatment system based on acupoint stimulation according to claim 5, characterized in that: The shortest path algorithm is based on the formula: d[v]=d[u]+αw(u,v)+βe(u,v)+γs(u,v) calculates the shortest path distance between acupoints, where d[v] is the shortest path distance from the source acupoint to acupoint v, d[u] is the known shortest path distance from the source acupoint to acupoint u, w(u,v) represents the basic treatment effect weight between node u and node v, e(u,v) represents the complementary treatment effect enhancement factor between acupoints u and v, s(u,v) represents the patient satisfaction influencing factor, α is the adjustment coefficient of the basic treatment effect weight, β is the adjustment coefficient of the complementary treatment effect enhancement factor, and γ is the adjustment coefficient of the patient satisfaction influencing factor.

7. The rehabilitation treatment system based on acupoint stimulation according to claim 1, characterized in that: The stimulation parameter optimization module comprises: The parameter simulation submodule performs acupoint stimulation parameter simulation based on the acupoint stimulation suggested path, wherein the parameter simulation includes multiple stimulation depths, frequencies and durations, evaluates the potential impact of differentiated parameter combinations on the treatment effect, and obtains acupoint stimulation simulation results; The optimization and adjustment submodule optimizes the acupoint stimulation parameters based on the acupoint stimulation simulation results by using a genetic algorithm, captures the parameter configuration of the target treatment effect by iteratively optimizing the parameter combination, and obtains the optimized parameter configuration result; The parameter setting submodule sets and adjusts the acupoint stimulation parameters based on the optimization parameter configuration result to match the expected treatment effect, optimize the efficiency and results of the rehabilitation treatment process, and obtain an improved acupoint stimulation strategy.

8. The rehabilitation treatment system based on acupoint stimulation according to claim 7, characterized in that: The genetic algorithm, according to the formula: Calculate the acupoint stimulation parameters, where is the parameter value of individual i in the next generation, is the parameter value of individual i in the current generation, is the parameter value of the best performing individual in the current generation, is the parameter value of the first randomly selected individual in the current generation, is the parameter value of the second individual randomly selected in the current generation, F is the scaling factor, and λ is the adjustment factor.

Images