A simulation training system and method for acupuncture needle manipulation

Abstract

The present invention relates to the technical field of traditional Chinese medicine acupuncture models, and specifically to a simulation training system and method for acupuncture needle manipulation. The present invention first adjusts the initial difference metric based on the posture difference influence index corresponding to the node to be analyzed of each two historical acupuncture action frame data to obtain a trajectory difference metric; obtains the participation metric of the node to be analyzed in the historical acupuncture action frame data according to the change in the moving distance of the node to be analyzed in the historical acupuncture action frame data; further clusters all historical acupuncture action frame data based on the distance metric value to obtain each acupuncture manipulation category; obtains a simulation training model of acupuncture needle manipulation based on all acupuncture manipulation categories. The present invention divides each acupuncture manipulation category more accurately and improves the accuracy of the simulation training model of acupuncture needle manipulation by fully considering the influence of posture differences on acupuncture manipulation and the degree of participation of different hand nodes in the acupuncture process.

Description

Technical Field

[0001] The present invention relates to the technical field of traditional Chinese medicine acupuncture models, and in particular to a simulation training system and method for acupuncture needle manipulation. Background Art

[0002] Acupuncture, a vital component of traditional Chinese medicine, encompasses two major areas: needling and moxibustion. Acupuncture involves inserting fine needles into the patient's body at specific acupuncture points, using twisting and lifting techniques to treat illnesses. To provide personalized treatment based on the patient's specific condition, acupuncturists must master various acupuncture techniques. Trainees can use the acupuncture virtual simulation training system to conduct simulation training in acupuncture techniques and refine their skills. To ensure the adequacy and effectiveness of simulation training, it is crucial to have a rich and well-classified historical acupuncture action frame data as support. Scientific classification of historical acupuncture action frame data is crucial for optimizing simulation training models. By carefully distinguishing between different data categories, the training model can flexibly adjust training parameters and strategies based on the data's characteristics, thereby accurately simulating real-world acupuncture scenarios.

[0003] Considering the differences in motion characteristics of different acupuncture techniques, existing techniques classify historical acupuncture motion frame data based on the differences in movement distances corresponding to each hand node in the historical acupuncture motion frame data. This classification fails to fully consider the impact of differences in acupuncture subject posture on acupuncture techniques and the varying degrees of involvement of different hand nodes in the acupuncture process. This makes it difficult to accurately classify historical acupuncture motion frame data, resulting in unreliable simulation training of acupuncture techniques. Summary of the Invention

[0004] In order to solve the technical problem that it is difficult to accurately construct a simulation training model for acupuncture needle manipulation in the prior art, the purpose of the present invention is to provide a simulation training system and method for acupuncture needle manipulation. The technical solutions adopted are as follows:

[0005] A simulation training system for acupuncture needle manipulation, comprising:

[0006] A data acquisition module is used to obtain the time series data of the movement distance corresponding to each hand node in each historical acupuncture action frame data;

[0007] A distance measurement module is used to obtain an initial difference measurement of the nodes to be analyzed of each two historical acupuncture action frame data according to the difference of the movement distance time series data corresponding to the nodes to be analyzed of each two historical acupuncture action frame data; obtain a posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data according to the period of the movement distance time series data corresponding to the nodes to be analyzed of each two historical acupuncture action frame data; adjust the initial difference measurement according to the posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data to obtain a trajectory difference measurement of the nodes to be analyzed of each two historical acupuncture action frame data; obtain a participation measurement of the node to be analyzed in the historical acupuncture action frame data according to the change of the movement distance of the node to be analyzed in the historical acupuncture action frame data; and obtain a distance measurement value of each two historical acupuncture action frame data by comprehensively analyzing the trajectory difference measurement of all the hand nodes of each two historical acupuncture action frame data and the participation measurement of the hand nodes in the historical acupuncture action frame data;

[0008] The simulation module is used to cluster all the historical acupuncture action frame data according to the distance measurement value between each two historical acupuncture action frame data to obtain each acupuncture technique category; and obtain a simulation training model of acupuncture acupuncture technique based on all acupuncture technique categories.

[0009] Furthermore, the method for obtaining the initial difference metric includes:

[0010] The DTW distance between the nodes to be analyzed of each two historical acupuncture action frame data and the movement distance time series data is used as the initial difference metric of the nodes to be analyzed of each two historical acupuncture action frame data.

[0011] Furthermore, the method for obtaining the body posture difference impact index includes:

[0012] Obtaining the action feature value corresponding to the node to be analyzed in the historical acupuncture action frame data according to the duration and amplitude of the period item of the movement distance time series data corresponding to the node to be analyzed in the historical acupuncture action frame data;

[0013] Obtaining a motion amplitude difference index corresponding to each of the nodes to be analyzed in each of the two historical acupuncture action frame data according to the absolute value of the difference between the motion feature values ​​corresponding to each of the nodes to be analyzed in each of the two historical acupuncture action frame data;

[0014] Obtaining a similarity index of the motion flow corresponding to each of the to-be-analyzed nodes of each of the two historical acupuncture action frames according to a correlation coefficient of the periodic item of the movement distance time series data corresponding to each of the to-be-analyzed nodes of each of the two historical acupuncture action frames;

[0015] According to the action process similarity index and the action amplitude difference index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data, the posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data is obtained.

[0016] Furthermore, the method for obtaining the action feature value includes:

[0017] For any periodic item of the movement distance time series data corresponding to the node to be analyzed of the historical acupuncture action frame data, the product of the duration and amplitude of the periodic item is calculated to obtain the action feature value corresponding to the node to be analyzed of the historical acupuncture action frame data.

[0018] Furthermore, the method for obtaining the body posture difference impact index includes:

[0019] The product of the action flow similarity index and the action amplitude difference index is calculated to obtain the posture difference influence index corresponding to each two nodes to be analyzed of the historical acupuncture action frame data.

[0020] Furthermore, the method for obtaining the participation metric includes:

[0021] Obtaining a first participation coefficient of the node to be analyzed in the historical acupuncture action frame data according to the correlation between the time series data of the movement distances corresponding to the node to be analyzed and all hand nodes in the historical acupuncture action frame data;

[0022] Calculate the mean of all movement distances corresponding to the node to be analyzed in the historical acupuncture action frame data to obtain the second participation coefficient of the node to be analyzed in the historical acupuncture action frame data;

[0023] The product of the first participation coefficient and the second participation coefficient is calculated and normalized to obtain a participation metric of the node to be analyzed in the historical acupuncture action frame data.

[0024] Furthermore, the method for obtaining the first participation coefficient includes:

[0025] For any historical acupuncture action frame data, calculate the absolute value of the Pearson correlation coefficient of the movement distance time series data corresponding to the node to be analyzed and the hand node to obtain the correlation coefficient between the node to be analyzed and the hand node;

[0026] The mean value of the correlation coefficient between the node to be analyzed and all hand nodes is calculated to obtain the first participation coefficient of the node to be analyzed in the historical acupuncture action frame data.

[0027] Furthermore, the distance measurement value acquisition method includes:

[0028] Calculating the absolute value of the difference between the participation metrics of the hand nodes of the two historical acupuncture action frame data in the historical acupuncture action frame data to obtain the participation difference metric of the hand nodes of the two historical acupuncture action frame data;

[0029] Calculating the mean of the participation metrics of the hand nodes of the two historical acupuncture action frame data to obtain the overall participation index of the hand nodes of the two historical acupuncture action frame data;

[0030] Calculate the product of the participation difference metric, the overall participation index, and the trajectory difference metric of the hand nodes of the two historical acupuncture action frame data to obtain the local difference degree of the hand nodes of the two historical acupuncture action frame data;

[0031] The cumulative sum of the local difference degrees of all hand nodes of two historical acupuncture action frame data is calculated to obtain the distance measurement value of each two historical acupuncture action frame data.

[0032] Furthermore, the method for obtaining the acupuncture technique category includes:

[0033] Using the K-Means clustering algorithm, all the historical acupuncture action frame data are clustered according to the distance measurement value between every two historical acupuncture action frame data to obtain various acupuncture technique categories.

[0034] A simulation training method for acupuncture needling techniques, wherein the method uses the above-mentioned simulation training system for acupuncture needling techniques to perform quality inspection.

[0035] The present invention has the following beneficial effects:

[0036] To accurately classify historical acupuncture action frame data, a distance metric must be constructed to accurately reflect the differences in acupuncture technique between two historical acupuncture action frames. First, an initial difference metric is constructed to reflect the difference in movement distance between the nodes to be analyzed. A posture difference impact index reflects the impact of patient posture differences on the acupuncture technique of the nodes to be analyzed. The initial difference metric is adjusted based on the posture difference impact index to produce a trajectory difference metric. The trajectory difference metric comprehensively considers the influence of movement distance and posture differences, and can more accurately reflect the differences in acupuncture technique corresponding to the nodes to be analyzed. Considering the varying degrees of participation of different hand nodes in the acupuncture process, nodes with higher participation have a greater impact on the acupuncture technique. The trajectory difference metric and the node participation metric of all hand nodes in each pair of historical acupuncture action frames are combined to calculate the distance metric between them. This distance metric fully accounts for the influence of patient posture differences on acupuncture technique and the degree of participation of different hand nodes in the acupuncture process, and can more accurately reflect the actual differences in acupuncture technique. The historical acupuncture action frame data can be more accurately divided into various acupuncture technique categories through distance measurement values, thereby improving the accuracy of the simulation training model of acupuncture technique. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 This is a system block diagram of a simulation training system for acupuncture needle manipulation provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features and effects of a simulation training system and method for acupuncture needle manipulation proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics of one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0041] The specific scheme of the simulation training system and method for acupuncture needle manipulation provided by the present invention is described in detail below with reference to the accompanying drawings.

[0042] The present invention provides a simulation training system and method for acupuncture needle manipulation. Figure 1 , which shows a system block diagram of a simulation training system for acupuncture needle manipulation provided by an embodiment of the present invention. The system includes: a data acquisition module 101, a distance measurement module 102, and a simulation module 103.

[0043] The data acquisition module 101 acquires the time series data of the movement distance corresponding to each hand node in each historical acupuncture action frame data.

[0044] In order to ensure the adequacy of simulation training, it is first necessary to obtain sufficient historical acupuncture action frame data as support; in order to classify the historical acupuncture action frame data, it is necessary to extract the movement distance time series data corresponding to each hand node from the historical acupuncture action frame data.

[0045] The detection system is used to obtain the data of each historical acupuncture action frame. Here is a brief description of the process of collecting a historical acupuncture action frame data:

[0046] In order to ensure that the details and three-dimensional information of the hand movements can be fully captured, the system installs cameras in three mutually perpendicular directions, such as the forward direction, the side direction and the upward direction. This multi-angle arrangement can effectively avoid the limitations of a single perspective and ensure the integrity and accuracy of the hand movements. The camera collects the acupuncturist's acupuncture process according to the preset sampling frequency, and obtains the original acupuncture images at each sampling moment. Since the collected original acupuncture images may be interfered with by noise, the system performs a noise reduction operation on the original acupuncture images to eliminate noise and external interference to ensure the accuracy of subsequent analysis. The image after noise reduction is called the acupuncture action frame image. The acupuncture action frame images collected by all cameras during the acupuncture process are counted to form historical acupuncture action frame data. In the present invention, each acupuncture process of the acupuncturist reflects a complete acupuncture technique process, and each historical acupuncture action frame data represents a complete acupuncture technique process.

[0047] Given that MediaPipe is a deep learning-based framework, it is particularly well-suited for gesture recognition and human posture analysis tasks. It analyzes each acupuncture action frame in real time, detecting and tracking key points of the hand and returning their position information, representing the key points as hand nodes. Using acupuncture action frames captured by three cameras from different directions, the system constructs a three-dimensional coordinate system for the hand nodes. For each hand node in each acupuncture action frame, the system obtains its spatial position in 3D space. For any historical acupuncture action frame, the system sequentially calculates the spatial position of the hand node in 3D space at each sampling moment, generating time series data for the corresponding spatial position of the hand node. For any hand node, the system calculates the Euclidean distance between each sampling moment and its corresponding spatial position at the previous sampling moment in the time series as the movement distance at that sampling moment. The system then calculates the movement distance at all sampling moments in chronological order, generating time series data for the movement distance of the hand node. This time series data reflects the movement changes of the hand node during the acupuncture process.

[0048] It should be noted that in the present invention, the same type of hand node refers to the key point of the same hand position. For example, in the historical acupuncture action frame data A, the node number of the thumb fingertip hand node is 1; in the historical acupuncture action frame data B, the node number of the thumb fingertip hand node is also 1; these two hand nodes belong to the same type of hand node, and they both represent the thumb fingertip. All historical acupuncture action frame data contain the same type of hand nodes. Such a design ensures the uniformity and comparability of the data, and provides a solid foundation for subsequent analysis and processing. In order to facilitate calculations, all indicator data involved in the calculations in the embodiment of the present invention are subjected to data preprocessing to eliminate the dimensionality effect. The specific means of removing the dimensionality effect are technical means well known to those skilled in the art and are not limited here.

[0049] The distance measurement module 102 is used to obtain the initial difference measurement of the nodes to be analyzed between each two historical acupuncture action frame data according to the difference in the moving distance time series data corresponding to the nodes to be analyzed between each two historical acupuncture action frame data; obtain the posture difference influence index corresponding to the nodes to be analyzed between each two historical acupuncture action frame data according to the period of the moving distance time series data corresponding to the nodes to be analyzed between each two historical acupuncture action frame data; adjust the initial difference measurement according to the posture difference influence index corresponding to the nodes to be analyzed between each two historical acupuncture action frame data to obtain the trajectory difference measurement of the nodes to be analyzed between each two historical acupuncture action frame data; obtain the participation measurement of the node to be analyzed in the historical acupuncture action frame data according to the change in the moving distance of the node to be analyzed in the historical acupuncture action frame data; and obtain the distance measurement value of each two historical acupuncture action frame data by combining the trajectory difference measurement of all hand nodes of each two historical acupuncture action frame data and the participation measurement of the hand nodes in the historical acupuncture action frame data.

[0050] To accurately classify historical acupuncture action frame data, a distance metric must be constructed to accurately reflect the differences in acupuncture technique between two historical acupuncture action frames. First, an initial difference metric is constructed to reflect the difference in movement distance between the nodes to be analyzed. A posture difference impact index reflects the impact of patient posture differences on the acupuncture technique of the nodes to be analyzed. The initial difference metric is adjusted based on the posture difference impact index to produce a trajectory difference metric. The trajectory difference metric comprehensively considers the influence of movement distance and posture differences, and can more accurately reflect the differences in acupuncture technique corresponding to the nodes to be analyzed. Considering the varying degrees of participation of different hand nodes in the acupuncture process, nodes with higher participation have a greater impact on the acupuncture technique. The trajectory difference metric and the node participation metric of all hand nodes in each pair of historical acupuncture action frames are combined to calculate the distance metric between them. This distance metric fully accounts for the influence of patient posture differences on acupuncture technique and the degree of participation of different hand nodes in the acupuncture process, and can more accurately reflect the actual differences in acupuncture technique.

[0051] In order to analyze the difference in acupuncture techniques between two historical acupuncture action frame data, firstly, the difference in the position movement of the same hand during the two historical acupuncture processes is locally analyzed, and any hand node is used as the node to be analyzed. In order to analyze the difference in the position movement of the node to be analyzed between the two historical acupuncture action frame data, preferably, in one embodiment of the present invention, the method for obtaining the initial difference metric includes:

[0052] The DTW distance between the time series data of the movement distances corresponding to the nodes to be analyzed between each two frames of historical acupuncture action data is used as the initial difference metric between the nodes to be analyzed between each two frames of historical acupuncture action data. It should be noted that the DTW distance is well known to those skilled in the art and can be obtained using the DTW (Dynamic Time Warping) algorithm, so this will not be discussed in detail here.

[0053] For the above steps, the historical acupuncture action frame data reflects the corresponding data of historical acupuncture techniques. Any hand node is used as the node to be analyzed. The DTW distance of the movement distance time series data corresponding to each two nodes to be analyzed in the historical acupuncture action frame data is used as the initial difference metric for each node to be analyzed in the historical acupuncture action frame data. A larger initial difference metric indicates a greater difference in the movement distance of the nodes to be analyzed in the two historical acupuncture action frame data, indicating a greater difference in the hand position movement corresponding to the nodes to be analyzed in the two historical acupuncture action frame data.

[0054] Considering that acupuncture techniques are usually cyclical, that is, there are multiple reciprocating identical movements, for example, the lifting and inserting method is a common acupuncture technique, which improves local or systemic function by slightly lifting and inserting the needle handle and repeatedly performing lifting and inserting movements. For the same acupuncture technique, differences in posture of different acupuncture subjects will lead to changes in acupuncture techniques. This change is usually reflected in the periodic changes in the amplitude of movement of the hand nodes. For example: for obese patients, the needle needs to be inserted deeper, and if the twisting method is used, the amplitude of movement is also larger. The movement amplitude of different hand nodes changes differently during the acupuncture process, so it is necessary to perform a detailed analysis of the changes in technique based on the hand nodes, in order to analyze the impact of the differences in posture of the acupuncture subjects corresponding to the two historical acupuncture processes on the movement of the nodes to be analyzed. Preferably, in one embodiment of the present invention, the method for obtaining the posture difference influence index includes:

[0055] Obtaining the action feature value corresponding to the node to be analyzed in the historical acupuncture action frame data according to the duration and amplitude of the periodic item of the movement distance time series data corresponding to the node to be analyzed;

[0056] Obtaining a motion amplitude difference index corresponding to the node to be analyzed in each of two historical acupuncture action frame data according to the absolute value of the difference between the motion feature values ​​corresponding to the node to be analyzed in each of two historical acupuncture action frame data;

[0057] Obtaining a similarity index of the motion flow corresponding to each of the to-be-analyzed nodes of each of the two historical acupuncture action frames based on the correlation coefficient of the periodic item of the moving distance time series data corresponding to each of the to-be-analyzed nodes of each of the two historical acupuncture action frames;

[0058] Based on the action process similarity index and action amplitude difference index corresponding to the nodes to be analyzed for each two historical acupuncture action frames, the posture difference influence index corresponding to the nodes to be analyzed for each two historical acupuncture action frames is obtained. It should be noted that the method for obtaining the periodic term is well known in the art of the present invention. The periodic term of the movement distance time series data corresponding to the nodes to be analyzed in the historical acupuncture action frames can be obtained by utilizing the STL (Seasonal-Trend Decomposition using LOESS) algorithm. The specific acquisition method is not detailed here. The method for obtaining the Pearson correlation coefficient is well known in the art and is not detailed here.

[0059] Specifically, for the periodic item of the moving distance time series data corresponding to the node to be analyzed of any historical acupuncture action frame data, the product of the duration and amplitude of the periodic item is calculated to obtain the action characteristic value corresponding to the node to be analyzed of the historical acupuncture action frame data; the absolute value of the difference between the action characteristic values ​​corresponding to the nodes to be analyzed of each two historical acupuncture action frame data is calculated to obtain the action amplitude difference index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data; the absolute value of the Pearson correlation coefficient of the periodic item of the moving distance time series data corresponding to the nodes to be analyzed of each two historical acupuncture action frame data is calculated to obtain the action process similarity index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data; the product of the action process similarity index and the action amplitude difference index is calculated to obtain the posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data. In one embodiment of the present invention, the formula of the posture difference influence index includes:

[0060]

[0061] Where, Indicates historical acupuncture action frame data and historical acupuncture action frame data Nodes to be analyzed Corresponding to the body posture difference impact index; Indicates historical acupuncture action frame data Nodes to be analyzed Corresponding action characteristic value; Indicates historical acupuncture action frame data Nodes to be analyzed Corresponding action characteristic value; Indicates historical acupuncture action frame data and historical acupuncture action frame data Nodes to be analyzed Pearson correlation coefficient of the periodic term of the time series data corresponding to the moving distance; Indicates historical acupuncture action frame data and historical acupuncture action frame data Nodes to be analyzed Corresponding action process similarity indicators; Indicates historical acupuncture action frame data and historical acupuncture action frame data Nodes to be analyzed Corresponding movement amplitude difference index; Indicates the absolute value symbol.

[0062] Following the above steps, a periodic term is extracted from the time-series data of the movement distance corresponding to the node to be analyzed in the historical acupuncture action frame data. The periodic term reflects the periodic variation characteristics of the acupuncture technique. For any node to be analyzed in the historical acupuncture action frame data, the product of the duration and amplitude of its periodic term is calculated to obtain the motion feature value of that node. The motion feature value, combined with the period length and amplitude of the acupuncture technique, comprehensively reflects the motion amplitude. For each pair of nodes to be analyzed in the historical acupuncture action frame data, the absolute value of the difference between their corresponding motion feature values ​​is calculated to obtain a motion amplitude difference index. The motion amplitude difference index reflects the variation in motion amplitude of the same node during different acupuncture processes, thereby reflecting the impact of body posture differences on motion amplitude. For each pair of nodes to be analyzed in the historical acupuncture action frame data, the absolute value of the Pearson correlation coefficient of the periodic term of the time-series data of the movement distance corresponding to them is calculated to obtain a motion process similarity index. The motion process similarity index reflects the similarity in the motion patterns of the two acupuncture action frame data. The larger the value of the motion process similarity index, the closer the motion patterns of the two acupuncture action frame data are, and the more likely they represent the same acupuncture technique. Taking into account that the more similar the acupuncture modes of two acupuncture action processes are and the greater the difference in movement amplitudes is, the greater the influence of posture differences on the acupuncture process is, and the more likely the two acupuncture processes represent the same acupuncture technique, the product of the movement process similarity index and the movement amplitude difference index is calculated to obtain the posture difference influence index corresponding to the to-be-analyzed nodes of each two historical acupuncture action frame data. The posture difference influence index represents the influence of posture differences on the acupuncture process.

[0063] In order to more accurately reflect the differences in acupuncture techniques corresponding to the nodes to be analyzed, preferably, in one embodiment of the present invention, the method for obtaining the trajectory difference metric includes:

[0064] Calculate the product of the negative correlation mapping result of the posture difference influence index corresponding to the node to be analyzed of each two historical acupuncture action frame data and the preset adjustment value to obtain the target adjustment value; calculate the product of the initial difference metric and the target adjustment value to obtain the trajectory difference metric of the node to be analyzed of each two historical acupuncture action frame data. It should be noted that the negative correlation mapping and technical means are well known to those skilled in the art. The negative correlation mapping can be in the form of inverse proportion or negative exponential power, which is not limited here. In one embodiment of the present invention, the preset adjustment value is 1.51, and the implementer can set it according to the implementation scenario.

[0065] For the above steps, negative correlation mapping is used to convert the posture difference influence index into a value that is negatively correlated with the initial difference metric. The larger the posture difference influence index, the greater the influence of the posture difference on the acupuncture technique. At this time, the weight of the initial difference metric should be reduced to avoid excessive influence of the posture difference on the trajectory difference metric. The preset adjustment value is used to adjust the amplitude of the negative correlation mapping result. The implementer can adjust the value according to the actual application scenario to optimize the calculation effect of the trajectory difference metric. The trajectory difference metric combines the initial difference metric and the posture difference influence index, and can more accurately reflect the difference in trajectory between the two acupuncture action frame data. The smaller the trajectory difference metric, the more similar the trajectories of the two acupuncture action frame data; the larger the trajectory difference metric, the greater the difference in the trajectories of the two acupuncture action frame data, that is, the greater the difference in acupuncture technique corresponding to the node to be analyzed.

[0066] In order to analyze the participation degree of hand nodes in the acupuncture process, preferably, in one embodiment of the present invention, the method for obtaining participation metrics includes:

[0067] Obtaining a first participation coefficient of the node to be analyzed in the historical acupuncture action frame data according to the correlation between the time series data of the movement distances corresponding to the node to be analyzed and all hand nodes in the historical acupuncture action frame data;

[0068] Calculate the mean of all movement distances corresponding to the node to be analyzed in the historical acupuncture action frame data to obtain the second participation coefficient of the node to be analyzed in the historical acupuncture action frame data;

[0069] The product of the first participation coefficient and the second participation coefficient is calculated and normalized to obtain the participation measure of the node to be analyzed in the historical acupuncture action frame data. In one embodiment of the present invention, the method for obtaining the first participation coefficient includes: for any historical acupuncture action frame data, calculating the absolute value of the Pearson correlation coefficient of the moving distance time series data corresponding to the node to be analyzed and the hand node, and obtaining the correlation coefficient between the node to be analyzed and the hand node; calculating the mean of the correlation coefficients between the node to be analyzed and all hand nodes, and obtaining the first participation coefficient of the node to be analyzed in the historical acupuncture action frame data. It should be noted that the normalization method is: using the norm normalization function for normalization, and limiting the numerical range to between 0 and 1. Normalization is a technical means well known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0070] For the above steps, the first participation coefficient reflects the correlation between the motion patterns of the node being analyzed and other hand nodes. A higher correlation coefficient indicates a stronger linkage between the node being analyzed and other nodes, and a higher degree of participation. The second participation coefficient reflects the amplitude of the motion of the node being analyzed during the acupuncture process. A larger amplitude indicates a greater influence of the node being analyzed on the acupuncture technique. The participation metric combines the first and second participation coefficients to accurately quantify the degree of involvement of the node being analyzed in the acupuncture process. A higher participation metric indicates a greater influence of the node on the acupuncture technique.

[0071] In order to more accurately reflect the actual differences in acupuncture techniques, preferably, in one embodiment of the present invention, the distance measurement value acquisition method includes:

[0072] Calculating the absolute value of the difference between the participation metrics of the hand nodes of the two historical acupuncture action frame data in the historical acupuncture action frame data to obtain the participation difference metric of the hand nodes of the two historical acupuncture action frame data;

[0073] Calculating the mean of the participation metrics of the hand nodes of the two historical acupuncture action frame data to obtain the overall participation index of the hand nodes of the two historical acupuncture action frame data;

[0074] Calculate the product of the participation difference metric, the overall participation index, and the trajectory difference metric of the hand nodes of the two historical acupuncture action frame data to obtain the local difference degree of the hand nodes of the two historical acupuncture action frame data;

[0075] The cumulative sum of the local difference degrees of all hand nodes of two historical acupuncture action frame data is calculated to obtain the distance measurement value of each two historical acupuncture action frame data.

[0076] For the above steps, for any two historical acupuncture action frames, we first focus on their hand nodes. Hand nodes play a vital role in the acupuncture process because they are the main part of performing the acupuncture action.

[0077] The absolute difference in participation metrics for each hand node in the two historical acupuncture action frames is calculated. A participation metric may refer to the hand node's activity level, position change, velocity, or other related metrics in a specific frame. By calculating the absolute difference, a participation difference metric is obtained for the hand nodes in the two historical acupuncture action frames, reflecting the subtle differences in their acupuncture movements. Next, the average of the participation metrics for the hand nodes in the two historical acupuncture action frames is calculated to obtain an overall participation metric. The overall participation metric represents the average activity level or importance of the hand node in the overall acupuncture action. With the participation difference metric and the overall participation metric, as well as the previously calculated trajectory difference metric, these three metrics are multiplied together to obtain the local difference between the hand nodes in the two historical acupuncture action frames. This product comprehensively considers the differences in the hand nodes' movements, overall activity level, and trajectory differences, providing a more comprehensive difference metric. Finally, the cumulative sum of the local difference between all hand nodes in the two historical acupuncture action frames is calculated to obtain a distance metric for each pair of historical acupuncture action frames. This cumulative sum represents the overall difference between the two historical acupuncture action frame data and is the basis for subsequent cluster analysis.

[0078] The simulation module 103 is used to cluster all historical acupuncture action frame data according to the distance measurement value between each two historical acupuncture action frame data to obtain each acupuncture technique category; and obtain a simulation training model of acupuncture acupuncture technique based on all acupuncture technique categories.

[0079] The historical acupuncture action frame data can be more accurately divided into various acupuncture technique categories through distance measurement values, thereby improving the accuracy of the simulation training model of acupuncture technique.

[0080] In order to more accurately classify the acupuncture technique categories, preferably, in one embodiment of the present invention, the method for obtaining the acupuncture technique categories includes:

[0081] Using the K-Means clustering algorithm, all historical acupuncture action frame data are clustered according to the distance measurement value between every two historical acupuncture action frame data to obtain the categories of various acupuncture techniques.

[0082] It should be noted that the K-Means clustering algorithm is a well-known technical means for those skilled in the art and will not be described in detail here. Only a brief description of the process of clustering all historical acupuncture action frame data based on the distance measurement value between each two historical acupuncture action frame data using the K-Means clustering algorithm in one embodiment of the present invention to obtain various acupuncture technique categories is provided:

[0083] Taking into account that the greater the distance measurement value between two historical acupuncture action frame data, the greater the difference in the gesture action, the optimal k value is obtained by the elbow method, and the distance measurement value of each two historical acupuncture action frame data is substituted into the clustering process. All historical acupuncture action frame data in the historical database are clustered to obtain clustering results and various acupuncture technique categories. The difference between historical acupuncture action frame data in different acupuncture technique categories is made as large as possible, and the difference between historical acupuncture action frame data in the same acupuncture technique category is made as small as possible. The similarity of acupuncture behavior of historical acupuncture action frame data in each acupuncture technique category is made as large as possible. In other embodiments of the present invention, the optimal k value of the K-Means clustering algorithm can be determined by other methods such as the silhouette coefficient method, which is not limited here.

[0084] Furthermore, based on the characteristics of each acupuncture technique category, the system constructs a simulation training model for acupuncture techniques; for example, for one acupuncture technique category, all historical acupuncture action frame data corresponding to the motion trajectory are used as the simulation training standard action for that acupuncture technique category; the simulation training model can use VR technology or other simulation means to show the user the standard action of each acupuncture technique for the user to conduct simulation training.

[0085] A simulation training method for acupuncture needling techniques, wherein the method uses any one of the above simulation training systems for acupuncture needling techniques to perform quality inspection.

[0086] In summary, the embodiment of the present invention provides a simulation training system and method for acupuncture needle manipulation. First, the initial difference metric is adjusted according to the posture difference influence index corresponding to the node to be analyzed of each two historical acupuncture action frame data to obtain the trajectory difference metric; according to the change in the moving distance of the node to be analyzed in the historical acupuncture action frame data, the participation metric of the node to be analyzed in the historical acupuncture action frame data is obtained; further, according to the distance metric value, all historical acupuncture action frame data are clustered to obtain each acupuncture manipulation category; according to all acupuncture manipulation categories, a simulation training model of acupuncture needle manipulation is obtained. The present invention divides each acupuncture manipulation category more accurately and improves the accuracy of the simulation training model of acupuncture needle manipulation by fully considering the influence of posture differences on acupuncture manipulation and the degree of participation of different hand nodes in the acupuncture process.

[0087] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A simulation training system for acupuncture needle manipulation, characterized in that: The system comprises: A data acquisition module is used to obtain the time series data of the movement distance corresponding to each hand node in each historical acupuncture action frame data; A distance measurement module is used to obtain an initial difference measurement of the nodes to be analyzed of each two historical acupuncture action frame data according to the difference of the movement distance time series data corresponding to the nodes to be analyzed of each two historical acupuncture action frame data; obtain a posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data according to the period of the movement distance time series data corresponding to the nodes to be analyzed of each two historical acupuncture action frame data; adjust the initial difference measurement according to the posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data to obtain a trajectory difference measurement of the nodes to be analyzed of each two historical acupuncture action frame data; obtain a participation measurement of the node to be analyzed in the historical acupuncture action frame data according to the change of the movement distance of the node to be analyzed in the historical acupuncture action frame data; and obtain a distance measurement value of each two historical acupuncture action frame data by comprehensively analyzing the trajectory difference measurement of all the hand nodes of each two historical acupuncture action frame data and the participation measurement of the hand nodes in the historical acupuncture action frame data; A simulation module is used to cluster all the historical acupuncture action frame data according to the distance measurement value between each two historical acupuncture action frame data to obtain each acupuncture technique category; and obtain a simulation training model of acupuncture acupuncture technique based on all acupuncture technique categories; The method for obtaining the body posture difference impact index includes: Obtaining the action feature value corresponding to the node to be analyzed in the historical acupuncture action frame data according to the duration and amplitude of the period item of the movement distance time series data corresponding to the node to be analyzed in the historical acupuncture action frame data; Obtaining a motion amplitude difference index corresponding to each of the nodes to be analyzed in each of the two historical acupuncture action frame data according to the absolute value of the difference between the motion feature values ​​corresponding to each of the nodes to be analyzed in each of the two historical acupuncture action frame data; Obtaining a similarity index of the motion flow corresponding to each of the to-be-analyzed nodes of each of the two historical acupuncture action frames according to a correlation coefficient of the periodic item of the movement distance time series data corresponding to each of the to-be-analyzed nodes of each of the two historical acupuncture action frames; According to the action process similarity index and the action amplitude difference index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data, the posture difference influence index corresponding to the nodes to be analyzed of each two historical acupuncture action frame data is obtained.

2. The acupuncture technique simulation training system according to claim 1, characterized in that: The method for obtaining the initial difference metric includes: The DTW distance between the nodes to be analyzed of each two historical acupuncture action frame data and the movement distance time series data is used as the initial difference metric of the nodes to be analyzed of each two historical acupuncture action frame data.

3. The acupuncture manipulation simulation training system according to claim 1, characterized in that: The method for obtaining the action characteristic value includes: For any periodic item of the movement distance time series data corresponding to the node to be analyzed of the historical acupuncture action frame data, the product of the duration and amplitude of the periodic item is calculated to obtain the action feature value corresponding to the node to be analyzed of the historical acupuncture action frame data.

4. The acupuncture manipulation simulation training system according to claim 1, characterized in that: The method for obtaining the body posture difference impact index includes: The product of the action flow similarity index and the action amplitude difference index is calculated to obtain the posture difference influence index corresponding to each two nodes to be analyzed of the historical acupuncture action frame data.

5. The acupuncture manipulation simulation training system according to claim 1, characterized in that: The method for obtaining the participation metric includes: Obtaining a first participation coefficient of the node to be analyzed in the historical acupuncture action frame data according to the correlation between the time series data of the movement distances corresponding to the node to be analyzed and all hand nodes in the historical acupuncture action frame data; Calculate the mean of all movement distances corresponding to the node to be analyzed in the historical acupuncture action frame data to obtain the second participation coefficient of the node to be analyzed in the historical acupuncture action frame data; The product of the first participation coefficient and the second participation coefficient is calculated and normalized to obtain a participation metric of the node to be analyzed in the historical acupuncture action frame data.

6. The acupuncture manipulation simulation training system according to claim 5, characterized in that: The method for obtaining the first participation coefficient includes: For any historical acupuncture action frame data, calculate the absolute value of the Pearson correlation coefficient of the movement distance time series data corresponding to the node to be analyzed and the hand node to obtain the correlation coefficient between the node to be analyzed and the hand node; The mean value of the correlation coefficient between the node to be analyzed and all hand nodes is calculated to obtain the first participation coefficient of the node to be analyzed in the historical acupuncture action frame data.

7. The acupuncture manipulation simulation training system according to claim 1, characterized in that: The distance measurement value acquisition method includes: Calculating the absolute value of the difference between the participation metrics of the hand nodes of the two historical acupuncture action frame data in the historical acupuncture action frame data to obtain the participation difference metric of the hand nodes of the two historical acupuncture action frame data; Calculating the mean of the participation metrics of the hand nodes of the two historical acupuncture action frame data to obtain the overall participation index of the hand nodes of the two historical acupuncture action frame data; Calculate the product of the participation difference metric, the overall participation index, and the trajectory difference metric of the hand nodes of the two historical acupuncture action frame data to obtain the local difference degree of the hand nodes of the two historical acupuncture action frame data; The cumulative sum of the local difference degrees of all hand nodes of two historical acupuncture action frame data is calculated to obtain the distance measurement value of each two historical acupuncture action frame data.

8. The acupuncture manipulation simulation training system according to claim 1, characterized in that: The method for obtaining the acupuncture technique category includes: Using the K-Means clustering algorithm, all the historical acupuncture action frame data are clustered according to the distance measurement value between every two historical acupuncture action frame data to obtain various acupuncture technique categories.

9. A simulation training method for acupuncture needle manipulation, characterized in that: The method uses a simulation training system for acupuncture needle manipulation as described in any one of claims 1 to 8 to perform quality inspection.

Images