Traditional Chinese Medicine Acupuncture Decision-making Assistance Method and System

Abstract

The embodiment of the present invention proposes a method and system for assisting decision-making in traditional Chinese medicine acupuncture, which relates to the technical field of traditional Chinese medicine acupuncture. The method uses a deep learning syndrome differentiation model that combines deep learning algorithms, traditional Chinese medicine syndrome differentiation theory and clinical cases to achieve high-precision syndrome type recognition, thereby improving the pertinence and accuracy of acupuncture treatment. And based on the syndrome type and symptom information of the patient, personalized acupuncture prescriptions are automatically generated to reduce the workload of clinicians and improve the efficiency and quality of acupuncture prescription formulation. The acupuncture prescription is then displayed using a three-dimensional human body model of the patient so that clinicians can intuitively view the acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction, thereby reducing the difficulty and error of acupuncture point positioning. Through accurate acupuncture point positioning and personalized acupuncture prescriptions, the effect of acupuncture treatment and patient satisfaction are improved, thereby improving patient experience.

Description

Technical Field

[0001] The present invention relates to the technical field of traditional Chinese medicine acupuncture, and in particular to a traditional Chinese medicine acupuncture decision-making assistance method and system. Background Art

[0002] In the prior art, a virtual human body model is usually generated according to the patient's body posture, and the virtual acupuncture points in the virtual human body model are converted into the patient's real body to achieve acupuncture treatment for the patient. However, there is still the problem of affecting the patient's recovery due to inaccurate diagnosis or inaccurate acupuncture techniques. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a method and system for assisting decision-making in traditional Chinese medicine acupuncture, which can significantly improve the accuracy of acupuncture treatment and the patient experience.

[0004] In order to achieve the above purpose, the technical solution adopted by the embodiment of the present invention is as follows:

[0005] In a first aspect, the present invention provides a method for assisting decision-making in traditional Chinese medicine and acupuncture, which is applied to a system for assisting decision-making in traditional Chinese medicine and acupuncture. The system for assisting decision-making in traditional Chinese medicine and acupuncture includes a collection module, a syndrome differentiation module, a prescription generation module, an acupoint display module, and a user interaction interface; the method includes:

[0006] The acquisition module transmits the patient's symptom information to the syndrome differentiation module; the symptom information includes the disease name and symptom information;

[0007] The syndrome differentiation module uses a deep learning syndrome differentiation model to analyze the disease information to obtain the patient's syndrome type; the deep learning syndrome differentiation model is obtained based on TCM syndrome differentiation theory and clinical case training;

[0008] The prescription generation module generates an acupuncture prescription according to the syndrome type and the symptom information; the acupuncture prescription includes acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction;

[0009] The acupuncture point display module marks the acupuncture point combination, the acupuncture point position, the acupuncture depth and the acupuncture direction on the three-dimensional human body model of the patient; the patient corresponds to the three-dimensional human body model one by one;

[0010] The user interaction interface displays the syndrome type, the acupuncture prescription and the three-dimensional human body model so that a clinician can perform acupuncture treatment on the patient.

[0011] In an optional embodiment, the syndrome differentiation module uses a deep learning syndrome differentiation model to analyze the symptom information to obtain the syndrome type of the patient, including:

[0012] The syndrome differentiation module pre-processes the symptom information to obtain standard symptom data;

[0013] The syndrome differentiation module extracts features from the standard symptom data to obtain a symptom vector;

[0014] The syndrome differentiation module inputs the symptom vector into the deep learning syndrome differentiation model, performs syndrome differentiation analysis, and obtains the syndrome type of the patient.

[0015] In an optional embodiment, the deep learning syndrome differentiation model includes an input layer, a hidden layer, an attention mechanism layer and an output layer; the hidden layer includes a multi-layer convolutional neural network and a recurrent neural network; the syndrome differentiation module inputs the symptom vector into the deep learning syndrome differentiation model, performs syndrome differentiation analysis, and obtains the patient's syndrome type, including:

[0016] The syndrome differentiation module transmits the received symptom vector to the multi-layer convolutional neural network using the input layer;

[0017] The syndrome differentiation module uses the multi-layer convolutional neural network to perform local feature extraction and pooling processing on the symptom vector to obtain a symptom feature map;

[0018] The syndrome differentiation module uses the recurrent neural network to extract the temporal relationship between symptoms according to the symptom feature graph to obtain the symptom temporal feature;

[0019] The syndrome differentiation module uses the attention mechanism layer to perform weighted processing on the symptom time series features using the attention mechanism to obtain symptom weighted features;

[0020] The syndrome differentiation module uses the output layer to perform multi-classification processing on the weighted features of the symptoms to obtain the syndrome type of the patient.

[0021] In an optional embodiment, the deep learning dialectical model is trained in the following manner:

[0022] The syndrome differentiation module extracts features from each symptom training information and the syndrome type corresponding to the symptom training information to obtain a symptom training vector and a theoretical syndrome type vector corresponding to each symptom training information; the symptom training information is obtained by preprocessing the symptom description and tongue and pulse condition in the TCM syndrome differentiation theory and clinical cases;

[0023] The syndrome differentiation module inputs the symptom training vector and the theoretical syndrome type vector into the first deep learning network to obtain the measured syndrome type corresponding to the symptom training information;

[0024] The syndrome differentiation module extracts vectors from the measured syndrome types to obtain measured syndrome type vectors;

[0025] The syndrome differentiation module iteratively updates the parameters and hyperparameters of the first deep learning network according to the loss information of the theoretical syndrome type vector and the measured syndrome type vector to obtain the deep learning syndrome differentiation model.

[0026] In an optional implementation, the prescription generation module generates an acupuncture prescription according to the syndrome type and the symptom information, including:

[0027] The prescription generation module pre-processes the symptom information to obtain standard symptom data;

[0028] The prescription generation module performs feature extraction on the standard symptom data and the syndrome type to obtain a symptom vector and a syndrome type vector;

[0029] The prescription generation module inputs the symptom vector, the syndrome type vector and the TCM knowledge base into a prescription generation model to obtain an acupuncture prescription for the patient; the TCM knowledge base includes acupuncture treatment rules and acupoint combinations.

[0030] In an optional embodiment, the prescription generation model is trained in the following manner:

[0031] The prescription generation module extracts features from each symptom training information, the syndrome type corresponding to the symptom training information, and the acupuncture point information corresponding to the symptom training information, to obtain a symptom training vector, a theoretical syndrome type vector, and an acupuncture point vector corresponding to each symptom training information; the symptom training information is obtained by pre-processing the symptom description and tongue and pulse images in the TCM syndrome differentiation theory and clinical cases;

[0032] The prescription generation module inputs the symptom training vector, the theoretical syndrome vector and the acupuncture point vector into a second deep learning network to obtain a training prescription corresponding to the symptom training information;

[0033] The prescription generation module extracts a vector from the training prescription to obtain a training prescription vector;

[0034] The prescription generation module iteratively updates the parameters and hyperparameters of the second deep learning network according to the acupuncture point vector and the loss information of the training prescription to obtain the prescription generation model.

[0035] In an optional embodiment, the three-dimensional human body model is constructed by:

[0036] The acupoint display module obtains the patient's human anatomy data and acupoint data;

[0037] The acupuncture point display module inputs the human anatomical data into a three-dimensional modeling tool to generate a three-dimensional original model of the patient;

[0038] The acupoint display module adds the acupoint data to the three-dimensional original model and renders the three-dimensional original model to obtain the three-dimensional human body model of the patient.

[0039] In an optional embodiment, the method further comprises:

[0040] The user interaction interface transmits the prescription modification information to the prescription generation module;

[0041] The prescription generating module updates the acupuncture prescription according to the prescription modification information to obtain an updated acupuncture prescription;

[0042] The acupoint display module re-marks the three-dimensional human body model of the patient according to the updated acupuncture prescription.

[0043] In a second aspect, the present invention provides a TCM acupuncture decision-making support system, which includes a collection module, a syndrome differentiation module, a prescription generation module, an acupoint display module, and a user interaction interface;

[0044] The acquisition module is used to transmit the patient's symptom information to the syndrome differentiation module; the symptom information includes the disease name and symptom information;

[0045] The syndrome differentiation module is used to analyze the disease information using a deep learning syndrome differentiation model to obtain the syndrome type of the patient; the deep learning syndrome differentiation model is obtained based on TCM syndrome differentiation theory and clinical case training;

[0046] The prescription generation module is used to generate an acupuncture prescription according to the syndrome type and the symptom information; the acupuncture prescription includes acupuncture point combination, acupoint location, acupuncture depth and acupuncture direction;

[0047] The acupuncture point display module is used to mark the acupuncture point combination, the acupuncture point position, the acupuncture depth and the acupuncture direction on the three-dimensional human body model of the patient; the patient corresponds to the three-dimensional human body model one by one;

[0048] The user interaction interface is used to display the syndrome type, the acupuncture prescription and the three-dimensional human body model so that the clinician can perform acupuncture treatment on the patient.

[0049] In an optional embodiment, the syndrome differentiation module is used to pre-process the symptom information to obtain standard symptom data;

[0050] The syndrome differentiation module is used to extract features from the standard syndrome data to obtain a syndrome vector;

[0051] The syndrome differentiation module is used to input the symptom vector into the deep learning syndrome differentiation model, perform syndrome differentiation analysis, and obtain the patient's syndrome type.

[0052] Compared with the prior art, the TCM acupuncture decision-making assistance method and system provided in the embodiment of the present invention utilizes a deep learning syndrome differentiation model that combines deep learning algorithms, TCM syndrome differentiation theory, and clinical cases to achieve high-precision syndrome type recognition, thereby improving the pertinence and accuracy of acupuncture treatment. And automatically generate personalized acupuncture prescriptions based on the patient's syndrome type and symptom information, reducing the workload of clinicians and improving the efficiency and quality of acupuncture prescription formulation. The acupuncture prescription is then displayed using a three-dimensional human body model of the patient so that clinicians can intuitively view the acupuncture point combination, acupuncture point location, acupuncture depth, and acupuncture direction, thereby reducing the difficulty and error of acupuncture point positioning. Through accurate acupuncture point positioning and personalized acupuncture prescriptions, the effect of acupuncture treatment and patient satisfaction are improved, thereby improving the patient experience.

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 A schematic flow chart of a method for assisting decision-making in traditional Chinese medicine and acupuncture provided in an embodiment of the present invention is shown.

[0056] Figure 2 Another schematic flow chart of the TCM acupuncture decision-making assistance method provided by an embodiment of the present invention is shown.

[0057] Figure 3 Another schematic flow chart of the TCM acupuncture decision-making assistance method provided by an embodiment of the present invention is shown.

[0058] Figure 4 Another schematic flow chart of the TCM acupuncture decision-making assistance method provided by an embodiment of the present invention is shown.

[0059] Figure 5 A block diagram of a traditional Chinese medicine acupuncture decision-making assistance system provided by an embodiment of the present invention is shown.

[0060] Icons: 400-TCM acupuncture decision support system; 401-acquisition module; 402-syndrome differentiation module; 403-prescription generation module; 404-acupoint display module; 405-user interaction interface. DETAILED DESCRIPTION

[0061] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0062] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0063] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0064] The inventors have found that the existing TCM acupuncture assisted positioning focuses more on using virtual reality technology and human posture recognition technology to achieve accurate positioning of acupuncture points on the real patient's body. The main method is to use head-mounted display devices to capture images of the real environment and its three-dimensional structure information, and convert the positions of virtual acupuncture points to the real patient's body through human posture recognition modules, human model alignment modules and acupuncture point positioning modules. In the visual interface, the virtual human model and acupuncture points are operated by gestures or voice commands while seeing the real patient's body, but there are still problems of misdiagnosis or inaccurate acupuncture techniques.

[0065] Based on this, the TCM acupuncture decision-making assistance method and system provided by the embodiment of the present invention utilizes a deep learning syndrome differentiation model that combines deep learning algorithms, TCM syndrome differentiation theory, and clinical cases to achieve high-precision syndrome type recognition, thereby improving the pertinence and accuracy of acupuncture treatment. And automatically generate personalized acupuncture prescriptions based on the patient's syndrome type and symptom information, reducing the workload of clinicians and improving the efficiency and quality of acupuncture prescription formulation. The acupuncture prescription is then displayed using a three-dimensional human body model of the patient so that clinicians can intuitively view the acupuncture point combination, acupuncture point location, acupuncture depth, and acupuncture direction, thereby reducing the difficulty and error of acupuncture point positioning. Through accurate acupuncture point positioning and personalized acupuncture prescriptions, the effect of acupuncture treatment and patient satisfaction are improved, thereby improving the patient experience.

[0066] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0067] Please refer to Figure 1 , Figure 1 A schematic flow chart of a method for assisting decision-making in traditional Chinese medicine and acupuncture provided by an embodiment of the present invention is shown. The method is applied to a system for assisting decision-making in traditional Chinese medicine and acupuncture, and the system for assisting decision-making in traditional Chinese medicine and acupuncture includes a collection module, a syndrome differentiation module, a prescription generation module, an acupoint display module, and a user interaction interface. The method includes the following steps:

[0068] Step S10, the acquisition module transmits the patient's symptom information to the syndrome differentiation module; the symptom information includes the disease name and symptom information.

[0069] In the embodiment of the present invention, the clinician inputs the patient's symptom information through the user interaction interface, and the symptom information includes the disease name and symptom information, and the symptom information includes Western medical diagnosis information such as pain location, symptom description, tongue and pulse condition, Western medical examination and inspection, etc. The user interaction interface transmits the symptom information to the syndrome differentiation module through the acquisition module to provide basic data for syndrome differentiation analysis.

[0070] Step S20, the syndrome differentiation module uses a deep learning syndrome differentiation model to analyze the symptom information to obtain the patient's syndrome type; the deep learning syndrome differentiation model is obtained based on TCM syndrome differentiation theory and clinical case training.

[0071] In an embodiment of the present invention, the syndrome differentiation module uses a pre-trained deep learning syndrome differentiation model to perform intelligent analysis on the disease name and symptom information to obtain a syndrome differentiation result, that is, the patient's syndrome type.

[0072] Step S30, the prescription generation module generates an acupuncture prescription according to the syndrome type and symptom information; the acupuncture prescription includes acupuncture point combination, acupoint location, acupuncture depth and acupuncture direction.

[0073] In the embodiment of the present invention, the prescription generation module matches the corresponding acupuncture treatment information according to the symptom type and symptom information, and generates an acupuncture prescription based on the acupuncture treatment information, wherein the acupuncture treatment information includes acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction.

[0074] Step S40, the acupuncture point display module marks the acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction on the patient's three-dimensional human body model; the patient and the three-dimensional human body model correspond one to one.

[0075] In the embodiment of the present invention, the acupuncture display module creates a corresponding 3D human body model for each patient, and the 3D human body model supports multi-angle rotation and zoom viewing. The acupuncture display module accurately displays the acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction on the patient's 3D human body model based on the acupuncture prescription, and provides detailed acupuncture instructions.

[0076] Step S50: The user interaction interface displays the syndrome type, acupuncture prescription and three-dimensional human body model so that the clinician can perform acupuncture treatment on the patient.

[0077] In the embodiment of the present invention, an intuitive and easy-to-use operating interface, i.e., a user interaction interface, is provided to facilitate clinicians to input the patient's symptom information, view the patient's syndrome type, acupuncture prescription, and the acupuncture point combination, point location, acupuncture depth, and acupuncture direction marked on the three-dimensional human body model.

[0078] In summary, the TCM acupuncture decision-making assistance method provided by the embodiment of the present invention is applied to the TCM acupuncture decision-making assistance system, and uses a deep learning syndrome differentiation model that combines deep learning algorithms, TCM syndrome differentiation theory and clinical cases to achieve high-precision syndrome type recognition, thereby improving the pertinence and accuracy of acupuncture treatment. And based on the syndrome type and symptom information of the patient, a personalized acupuncture prescription is automatically generated to reduce the workload of clinicians and improve the efficiency and quality of acupuncture prescription formulation. The acupuncture prescription is then displayed using a three-dimensional human body model of the patient so that clinicians can intuitively view the acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction, thereby reducing the difficulty and error of acupuncture point positioning. Through accurate acupuncture point positioning and personalized acupuncture prescriptions, the effect of acupuncture treatment and patient satisfaction are improved, thereby improving patient experience.

[0079] Optionally, a possible implementation method is provided below for how to obtain the patient's symptom type. Figure 2 , Figure 1 The sub-steps of step S20 may include:

[0080] Step S201: the syndrome differentiation module pre-processes the symptom information to obtain standard symptom data.

[0081] In the embodiment of the present invention, the syndrome differentiation module performs standardization processing on the symptom information, so that the symptom information with the same meaning but different expressions is unified to obtain standard disease data.

[0082] Step S202: The syndrome differentiation module extracts features from the standard syndrome data to obtain a syndrome vector.

[0083] Step S203: The syndrome differentiation module inputs the symptom vector into the deep learning syndrome differentiation model to perform syndrome differentiation analysis to obtain the syndrome type of the patient.

[0084] In an embodiment of the present invention, the syndrome differentiation module can use word embedding technology to convert the text information of standard disease data into numerical features, that is, vector form (symptom vector). The syndrome differentiation module inputs the symptom vector into a pre-trained deep learning syndrome differentiation model, and instructs the deep learning syndrome differentiation model to analyze the corresponding syndrome type according to the symptom vector.

[0085] Optionally, in practical applications, the deep learning dialectical model includes an input layer, a hidden layer, an attention mechanism layer and an output layer; the hidden layer includes a multi-layer convolutional neural network and a recurrent neural network; as to how to use the deep learning dialectical model to obtain the patient's symptom type, a possible implementation method is provided below. Figure 2 The sub-steps of step S203 may include:

[0086] The syndrome differentiation module uses the input layer to transfer the received symptom vector to the multi-layer convolutional neural network; the syndrome differentiation module uses the multi-layer convolutional neural network to perform local feature extraction and pooling processing on the symptom vector to obtain the symptom feature map; the syndrome differentiation module uses the recurrent neural network to extract the temporal relationship between symptoms according to the symptom feature map to obtain the symptom temporal features; the syndrome differentiation module uses the attention mechanism layer to perform weighted processing on the symptom temporal features to obtain the symptom weighted features; the syndrome differentiation module uses the output layer to perform multi-classification processing on the symptom weighted features to obtain the patient's syndrome type.

[0087] In an embodiment of the present invention, the input layer is used to receive the symptom vector corresponding to the symptom information transmitted by the acquisition module. The symptom information includes pain location, symptom description, tongue and pulse condition, Western medicine examination and other Western medicine diagnosis information. The symptom information needs to be preprocessed before being transmitted to the input layer, and then the preprocessed data is converted into a vector form with clear numerical characteristics (i.e., the symptom vector), laying the foundation for dialectical analysis.

[0088] The hidden layer creatively adopts a hybrid structure that combines multi-layer convolutional neural networks (CNN) and recurrent neural networks (RNN). CNN, like a keen observer, can accurately extract local features in the disease vector. Whether it is complex tongue or pulse information, CNN can effectively capture it. RNN is good at capturing the subtle temporal relationship and contextual dependency between symptoms, making the deep learning dialectical model's understanding of symptom information more comprehensive and in-depth.

[0089] Furthermore, the convolutional neural network includes a convolutional layer and a pooling layer. The calculation formula of the convolutional layer is y=δ(W·x+b), where y is a high-dimensional feature map; x is a symptom vector; W is a convolution kernel, which is the core element of the convolution operation and is responsible for extracting local features in the symptom vector x; b is a bias term used to adjust the activation threshold of neurons; δ is an activation function that introduces nonlinear factors into the convolutional neural network, so that the deep learning dialectical model can handle complex nonlinear relationships.

[0090] The pooling layer can use the maximum pooling and average pooling methods to convert high-dimensional feature maps into disease feature maps. The pooling operation is like a data compressor. By summarizing local features (maximum pooling takes the maximum value, average pooling takes the average value), it effectively reduces the feature dimension, enhances the generalization ability of the model, avoids overfitting of the deep learning dialectical model, and enables the deep learning dialectical model to output accurate results more stably when facing new disease vectors.

[0091] The hidden state update formula in the recurrent neural network is h t =δ(W h ·h t-1 +W x ·x t +b). Among them, h t is the current hidden state, which is an internal representation of the network's input information at the current moment; h t-1 is the hidden state of the previous moment, which reflects the memory ability of the network and enables information to be transmitted in time series; W h is the weight corresponding to the hidden state at the previous moment, which is used to adjust the influence of the hidden state at the previous moment on the current hidden state; x t is the current input feature vector (i.e., the disease feature map); W x is the weight corresponding to the current input feature vector, which is used to adjust the influence of the input feature vector on the current hidden state.

[0092] The formula for calculating the attention weight at the attention mechanism layer is:

[0093]

[0094] Among them, α is the attention weight, e is the correlation score between the current symptom (i.e., the temporal characteristics of the current symptom) and other symptoms, which represents the importance of each symptom to the current symptom information; e i is the correlation score between the ith symptom and other symptoms; N is the number of symptoms (i.e., the number of temporal characteristics of the disease).

[0095] The weight of each symptom is obtained by the calculation formula of attention weight, and then the weighted summation formula is used to weight the time series characteristics of the symptoms to obtain the weighted characteristics of the symptoms. The weighted characteristics of the symptoms integrate the information of each symptom, so that the deep learning syndrome differentiation model can focus on key symptoms and further improve the accuracy of syndrome differentiation. The weighted summation formula is:

[0096]

[0097] Among them, c is the disease weighted feature; α i is the attention weight of the temporal feature of the ith disease; h i is the hidden state at the ith moment (i.e., the temporal characteristics of the disease).

[0098] The output layer uses the softmax function to perform multi-classification processing on the weighted features of the symptoms, and finally obtains the patient's syndrome type. This design combines the advantages of CNN and RNN, so that the deep learning syndrome differentiation model can not only keenly mine the local features of the symptoms, but also accurately extract the temporal relationship between the symptoms, greatly improving the accuracy of syndrome differentiation. In order to further improve the syndrome differentiation effect, an attention mechanism is also introduced, which can give different attention weights to different symptoms according to their importance, so as to focus on key symptoms more accurately and provide more favorable support for identifying syndrome types.

[0099] Alternatively, the following is a possible implementation method for how to train a deep learning dialectical model. Figure 3 , the Chinese medicine acupuncture decision-making assistance method also includes the following steps:

[0100] Step S60, the syndrome differentiation module extracts features from each symptom training information and the syndrome type corresponding to the symptom training information, and obtains a symptom training vector and a theoretical syndrome type vector corresponding to each symptom training information; the symptom training information is obtained after pre-processing the symptom descriptions and tongue and pulse images in the TCM syndrome differentiation theory and clinical cases.

[0101] In an embodiment of the present invention, TCM syndrome differentiation theory and clinical cases are collected. TCM syndrome differentiation theory includes Chinese medicinal materials, literature, and medical records of famous doctors. Case data include basic information of patients (such as age, gender), disease name, symptom description (such as pain location, degree, duration, accompanying symptoms, etc.), tongue and pulse condition, syndrome type, acupuncture prescriptions and other information.

[0102] The collected symptom descriptions and tongue and pulse images are cleaned and preprocessed to remove duplicate, incomplete or abnormal data. Symptom descriptions are standardized to unify symptom information with the same meaning but different expressions, and multiple symptom training information is obtained to ensure that each symptom training information is of high quality and provide a pure and reliable data foundation for subsequent model training. The syndrome type in the same case data is determined as the syndrome type corresponding to the symptom training information.

[0103] Word Embedding (WE) is used to convert symptom training information and syndrome types into numerical features to obtain symptom training vectors and theoretical syndrome type vectors, creating convenient conditions for the deep learning syndrome differentiation model to smoothly process and learn symptom training information and syndrome types.

[0104] Step S70: The syndrome differentiation module inputs the symptom training vector and the theoretical syndrome type vector into the first deep learning network to obtain the measured syndrome type corresponding to the symptom training information.

[0105] Step S80, the syndrome differentiation module extracts vectors from the measured syndrome types to obtain measured syndrome type vectors.

[0106] Step S90: The syndrome differentiation module iteratively updates the parameters and hyperparameters of the first deep learning network according to the loss information of the theoretical syndrome type vector and the measured syndrome type vector to obtain a deep learning syndrome differentiation model.

[0107] In an embodiment of the present invention, the syndrome differentiation module inputs the symptom training vector and the theoretical syndrome type vector into the first deep learning network, instructs the first deep learning network to perform a syndrome differentiation analysis on the symptom training vector based on the theoretical syndrome type vector, and obtains the measured syndrome type corresponding to the symptom training information; uses word embedding to convert the test syndrome type into a measured syndrome type vector, and optimizes the parameters and hyperparameters of the first deep learning network through the back propagation algorithm based on the loss information of the theoretical syndrome type vector and the measured syndrome type vector, to obtain a deep learning syndrome differentiation model, thereby improving the accuracy of the syndrome differentiation analysis.

[0108] Optionally, a possible implementation method is provided below for how to obtain the patient's acupuncture prescription. Figure 2 , Figure 1 The sub-steps of step S30 may include:

[0109] Step S301: The prescription generation module pre-processes the symptom information to obtain standard symptom data.

[0110] In the embodiment of the present invention, the prescription generation module performs cleaning preprocessing on the symptom information, and performs standardization processing on the processed symptom information to obtain standard symptom information.

[0111] Step S302: The prescription generation module extracts features from the standard symptom data and syndrome types to obtain symptom vectors and syndrome type vectors.

[0112] Step S303, the prescription generation module inputs the symptom vector, the syndrome type vector and the TCM knowledge base into the prescription generation model to obtain the patient's acupuncture prescription; the TCM knowledge base includes acupuncture treatment rules and acupoint combinations.

[0113] In the embodiment of the present invention, a TCM knowledge base is pre-established to store acupuncture treatment rules and acupoint combinations, so that the prescription generation model can learn acupuncture point combinations, acupuncture techniques and treatment principles corresponding to different syndrome types, etc. Acupuncture techniques include acupuncture depth and acupuncture direction.

[0114] The prescription generation model includes an input layer, a hidden layer and an output layer. The prescription generation module inputs the symptom vector, the syndrome type vector and the access address of the TCM knowledge base into the prescription generation model, and instructs the prescription generation model to analyze the symptom vector and the syndrome type vector in combination with the acupuncture treatment rules and acupoint compatibility in the TCM knowledge base.

[0115] Among them, the input layer receives the symptom vector, syndrome type vector and the access address of the TCM knowledge base and passes them to the hidden layer. The symptom vector and syndrome type vector are the Chinese medicine basis for generating acupuncture prescriptions. The hidden layer learns acupuncture treatment rules and acupoint compatibility through the access address of the TCM knowledge base, and analyzes the symptom vector and syndrome type vector. This deep integration of traditional TCM theory and modern machine learning algorithms provides a solid technical guarantee for generating personalized acupuncture prescriptions.

[0116] The output layer can output accurate and personalized acupuncture prescriptions, covering in detail key information such as acupoint combinations, acupoint locations, and acupuncture techniques, ensuring that the generated acupuncture prescriptions conform to both the algorithm logic of modern science and the principles of traditional Chinese medicine theory.

[0117] It can be seen that the embodiment of the present invention performs differential analysis based on the patient's symptom information and syndrome type, and automatically generates a personalized acupuncture prescription, so that the acupuncture prescription is more in line with the patient's condition, can be symptomatic acupuncture, and improve the treatment effect. At the same time, the prescription generation model further improves the accuracy of acupuncture prescriptions by continuously learning the acupuncture point combinations, acupuncture techniques and treatment principles corresponding to different syndrome types in the traditional Chinese medicine knowledge base.

[0118] Optionally, a possible implementation method for how to train the prescription generation model is provided below.

[0119] Please refer to Figure 4 , the Chinese medicine acupuncture decision-making assistance method also includes the following steps:

[0120] Step S100, the prescription generation module extracts features from each symptom training information, the syndrome type corresponding to the symptom training information, and the acupuncture point information corresponding to the symptom training information, to obtain a symptom training vector, a theoretical syndrome type vector, and an acupuncture point vector corresponding to each symptom training information; the symptom training information is obtained after pre-processing the symptom descriptions and tongue and pulse images in the TCM syndrome differentiation theory and clinical cases.

[0121] Step S110: The prescription generation module inputs the symptom training vector, the theoretical syndrome vector and the acupuncture point vector into the second deep learning network to obtain a training prescription corresponding to the symptom training information.

[0122] Step S120 : the prescription generation module extracts a vector from the training prescription to obtain a training prescription vector.

[0123] Step S130: The prescription generation module iteratively updates the parameters and hyperparameters of the second deep learning network according to the acupuncture point vector and the loss information of the training prescription to obtain a prescription generation model.

[0124] In an embodiment of the present invention, the prescription generation module trains the second deep learning network based on the symptom training vector, the theoretical syndrome vector and the acupuncture point vector, instructs the second deep learning network to generate a training prescription vector based on the symptom training vector and the theoretical syndrome vector, and then iteratively updates the parameters and hyperparameters of the second deep learning network based on the loss information of the acupuncture point vector and the training prescription. By continuously adjusting the hyperparameters, the best performance of the second deep learning network is explored to ensure that the prescription generation model can stably and efficiently output acupuncture prescriptions when facing various input data.

[0125] Optionally, a possible implementation method is provided below for how to construct a three-dimensional human body model. The Chinese medicine acupuncture decision-making assistance method also includes the following steps:

[0126] The acupoint display module obtains the patient's human anatomical data and acupoint data; the acupoint display module inputs the human anatomical data into the three-dimensional modeling tool to generate a three-dimensional original model of the patient; the acupoint display module adds the acupoint data to the three-dimensional original model, and renders the three-dimensional original model to obtain the patient's three-dimensional human body model.

[0127] In the embodiment of the present invention, the patient is scanned by a scanning device to obtain the patient's human anatomical data, which includes bones, muscles, blood vessels and nerves. The information such as the location, depth and direction of acupuncture points in traditional Chinese medicine is sorted out and mapped to the patient's human anatomical data to obtain acupuncture point data, so as to improve the accuracy of acupuncture points.

[0128] Based on human anatomical data, a realistic 3D model is created using 3D modeling tools (such as Blender and Maya) to obtain the original 3D model of the patient. According to the acupoint data in the acupuncture prescription, the location, acupuncture depth and direction of each acupoint are dynamically marked on the 3D human model to ensure that the location of each acupoint is accurate. Different colors or marking styles are supported to distinguish different acupoints to improve recognition.

[0129] The 3D original model is rendered using advanced 3D rendering technology to obtain the patient's 3D human body model, so as to improve the visual effect of the 3D human body model and make it closer to the real human body. It supports visual display under different lighting conditions, allowing users to view acupoint information on the patient's 3D human body model by clicking, dragging, zooming, rotating, etc. on the user interaction interface, so that users can observe the acupoint location from different angles.

[0130] It can be seen that the embodiment of the present invention cleverly integrates the acupoint display function and the user interaction interface, so that clinicians can easily view and modify the acupoint combination, acupoint location, acupuncture depth and acupuncture direction in the acupuncture prescription, and verify the accuracy of the acupoint location and acupuncture technique in real time.

[0131] Optionally, a possible implementation method is provided below for updating the clinician's modification suggestion for acupuncture prescription. The TCM acupuncture decision-making assistance method also includes the following steps:

[0132] The user interaction interface transmits the prescription modification information to the prescription generation module; the prescription generation module updates the acupuncture prescription according to the prescription modification information to obtain an updated acupuncture prescription; the acupoint display module re-marks the patient's three-dimensional human body model according to the updated acupuncture prescription.

[0133] In an embodiment of the present invention, the clinician can modify the automatically generated acupuncture prescription according to the actual situation of the patient, and transmit the prescription modification information to the prescription generation module through the user interaction interface. The prescription generation module updates the prescription modification information to the acupuncture prescription and saves the updated acupuncture prescription.

[0134] The acupoint display module marks the updated acupoint combination, acupoint location, acupuncture depth and acupuncture direction on the patient's 3D human model so that clinicians can verify the modified acupoint combination, acupoint location, acupuncture depth and acupuncture direction through the 3D human model. Clinicians perform treatment according to the finalized acupuncture prescription and collect treatment feedback through the TCM acupuncture decision-making support system for optimization and updating of the TCM acupuncture decision-making support system.

[0135] Based on the same inventive concept, the basic principles and technical effects of the TCM acupuncture decision support system provided by the embodiment of the present invention are the same as those of the above-mentioned embodiment. For the sake of brief description, for parts not mentioned in this embodiment, reference may be made to the corresponding contents in the above-mentioned embodiment.

[0136] Please refer to Figure 5 , Figure 5 A schematic block diagram of a TCM acupuncture decision support system 400 provided in an embodiment of the present invention. The TCM acupuncture decision support system 400 comprises a collection module 401, a syndrome differentiation module 402, a prescription generation module 403, an acupoint display module 404 and a user interaction interface 405.

[0137] The acquisition module 401 is used to transmit the patient's symptom information to the syndrome differentiation module; the symptom information includes the disease name and symptom information.

[0138] The syndrome differentiation module 402 is used to analyze the symptom information using a deep learning syndrome differentiation model to obtain the patient's syndrome type; the deep learning syndrome differentiation model is obtained based on TCM syndrome differentiation theory and clinical case training.

[0139] The prescription generation module 403 is used to generate an acupuncture prescription according to the syndrome type and symptom information; the acupuncture prescription includes acupuncture point combination, acupoint location, acupuncture depth and acupuncture direction.

[0140] The acupoint display module 404 is used to mark the acupoint combination, acupoint location, acupoint depth and acupuncture direction on the three-dimensional human body model of the patient; the patient corresponds to the three-dimensional human body model one by one.

[0141] The user interaction interface 405 is used to display syndrome types, acupuncture prescriptions and a three-dimensional human body model so that clinicians can perform acupuncture treatment on patients.

[0142] In summary, the TCM acupuncture decision-making support system provided by the embodiment of the present invention utilizes a deep learning syndrome differentiation model that combines deep learning algorithms, TCM syndrome differentiation theory, and clinical cases to achieve high-precision syndrome type recognition, thereby improving the pertinence and accuracy of acupuncture treatment. And based on the syndrome type and symptom information of the patient, personalized acupuncture prescriptions are automatically generated to reduce the workload of clinicians and improve the efficiency and quality of acupuncture prescription formulation. The acupuncture prescription is then displayed using a three-dimensional human body model of the patient so that clinicians can intuitively view the acupuncture point combination, acupuncture point location, acupuncture depth, and acupuncture direction, thereby reducing the difficulty and error of acupuncture point positioning. Through accurate acupuncture point positioning and personalized acupuncture prescriptions, the effect of acupuncture treatment and patient satisfaction are improved, thereby improving the patient experience.

[0143] Optionally, the syndrome differentiation module 402 is used to preprocess the symptom information to obtain standard symptom data; the syndrome differentiation module 402 is used to extract features from the standard symptom data to obtain a symptom vector; the syndrome differentiation module 402 is used to input the symptom vector into a deep learning syndrome differentiation model to perform syndrome differentiation analysis to obtain the patient's syndrome type.

[0144] Optionally, the deep learning syndrome differentiation model includes an input layer, a hidden layer, an attention mechanism layer and an output layer; the hidden layer includes a multi-layer convolutional neural network and a recurrent neural network; the syndrome differentiation module 402 is used to use the input layer to transfer the received disease vector to the multi-layer convolutional neural network; the syndrome differentiation module 402 is used to use the multi-layer convolutional neural network to perform local feature extraction and pooling processing on the disease vector to obtain a disease feature map; the syndrome differentiation module 402 is used to use the recurrent neural network to extract the temporal relationship between diseases according to the disease feature map to obtain the disease temporal features; the syndrome differentiation module 402 is used to use the attention mechanism layer to perform weighted processing on the disease temporal features to obtain the disease weighted features; the syndrome differentiation module 402 is used to use the output layer to perform multi-classification processing on the disease weighted features to obtain the patient's syndrome type.

[0145] Optionally, the syndrome differentiation module 402 is further used to extract features from each symptom training information and the syndrome type corresponding to the symptom training information, and obtain a symptom training vector and a theoretical syndrome type vector corresponding to each symptom training information; the symptom training information is obtained by preprocessing the symptom description and tongue and pulse images in the TCM syndrome differentiation theory and clinical cases;

[0146] The syndrome differentiation module 402 is also used to input the symptom training vector and the theoretical syndrome type vector into the first deep learning network to obtain the measured syndrome type corresponding to the symptom training information; extract the vector of the measured syndrome type to obtain the measured syndrome type vector; iteratively update the parameters and hyperparameters of the first deep learning network according to the loss information of the theoretical syndrome type vector and the measured syndrome type vector to obtain a deep learning syndrome differentiation model.

[0147] Optionally, the prescription generation module 403 is used to pre-process the symptom information to obtain standard symptom data; the prescription generation module 403 is used to extract features from the standard symptom data and syndrome types to obtain symptom vectors and syndrome type vectors; the prescription generation module 403 is used to input the symptom vectors, syndrome type vectors and the traditional Chinese medicine knowledge base into the prescription generation model to obtain the patient's acupuncture prescription; the traditional Chinese medicine knowledge base includes acupuncture treatment rules and acupoint combinations.

[0148] Optionally, the prescription generation module 403 is used to extract features from each symptom training information, the syndrome type corresponding to the symptom training information, and the acupuncture point information corresponding to the symptom training information, to obtain a symptom training vector, a theoretical syndrome type vector, and an acupuncture point vector corresponding to each symptom training information; the symptom training information is obtained after pre-processing the symptom descriptions and tongue and pulse images in the TCM syndrome differentiation theory and clinical cases.

[0149] The prescription generation module 403 is used to input the symptom training vector, theoretical syndrome vector and acupuncture point vector into the second deep learning network to obtain the training prescription corresponding to the symptom training information; the prescription generation module 403 is used to extract the vector of the training prescription to obtain the training prescription vector; the prescription generation module 403 is used to iteratively update the parameters and hyperparameters of the second deep learning network according to the loss information of the acupuncture point vector and the training prescription to obtain the prescription generation model.

[0150] Optionally, the acupoint display module 404 is used to obtain the patient's human anatomical data and acupoint data; the acupoint display module 404 is used to input the human anatomical data into a three-dimensional modeling tool to generate a three-dimensional original model of the patient; the acupoint display module 404 is used to add the acupoint data to the three-dimensional original model and render the three-dimensional original model to obtain a three-dimensional human body model of the patient.

[0151] Optionally, the user interaction interface 405 is used to transmit the prescription modification information to the prescription generation module; the prescription generation module 403 is used to update the acupuncture prescription according to the prescription modification information to obtain an updated acupuncture prescription; the acupoint display module 404 is used to re-mark the patient's three-dimensional human body model according to the updated acupuncture prescription.

[0152] In several embodiments provided by the present invention, it should be understood that the disclosed system and method can also be implemented in other ways. The system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0153] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0154] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0155] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for assisting decision making in traditional Chinese medicine and acupuncture, characterized in that: Applied to a Chinese medicine acupuncture decision-making support system, the Chinese medicine acupuncture decision-making support system includes a collection module, a syndrome differentiation module, a prescription generation module, an acupoint display module and a user interaction interface; the method includes: The acquisition module transmits the patient's symptom information to the syndrome differentiation module; the symptom information includes the disease name and symptom information; The syndrome differentiation module uses a deep learning syndrome differentiation model to analyze the disease information to obtain the patient's syndrome type; the deep learning syndrome differentiation model is obtained based on TCM syndrome differentiation theory and clinical case training; The prescription generation module generates an acupuncture prescription according to the syndrome type and the symptom information; the acupuncture prescription includes acupuncture point combination, acupuncture point location, acupuncture depth and acupuncture direction; The acupuncture point display module marks the acupuncture point combination, the acupuncture point position, the acupuncture depth and the acupuncture direction on the three-dimensional human body model of the patient; the patient corresponds to the three-dimensional human body model one by one; The user interaction interface displays the syndrome type, the acupuncture prescription and the three-dimensional human body model so that a clinician can perform acupuncture treatment on the patient.

2. The Chinese medicine acupuncture decision-making assistance method according to claim 1, characterized in that: The syndrome differentiation module uses a deep learning syndrome differentiation model to analyze the symptom information to obtain the patient's syndrome type, including: The syndrome differentiation module pre-processes the symptom information to obtain standard symptom data; The syndrome differentiation module extracts features from the standard symptom data to obtain a symptom vector; The syndrome differentiation module inputs the symptom vector into the deep learning syndrome differentiation model, performs syndrome differentiation analysis, and obtains the syndrome type of the patient.

3. The Chinese medicine acupuncture decision-making assistance method according to claim 2, characterized in that: The deep learning syndrome differentiation model includes an input layer, a hidden layer, an attention mechanism layer and an output layer; the hidden layer includes a multi-layer convolutional neural network and a recurrent neural network; the syndrome differentiation module inputs the symptom vector into the deep learning syndrome differentiation model, performs syndrome differentiation analysis, and obtains the syndrome type of the patient, including: The syndrome differentiation module transmits the received symptom vector to the multi-layer convolutional neural network using the input layer; The syndrome differentiation module uses the multi-layer convolutional neural network to perform local feature extraction and pooling processing on the symptom vector to obtain a symptom feature map; The syndrome differentiation module uses the recurrent neural network to extract the temporal relationship between symptoms according to the symptom feature graph to obtain the symptom temporal feature; The syndrome differentiation module uses the attention mechanism layer to perform weighted processing on the symptom time series features using the attention mechanism to obtain symptom weighted features; The syndrome differentiation module uses the output layer to perform multi-classification processing on the weighted features of the symptoms to obtain the syndrome type of the patient.

4. The Chinese medicine acupuncture decision-making assistance method according to claim 1, characterized in that: The deep learning dialectical model is trained in the following way: The syndrome differentiation module extracts features from each symptom training information and the syndrome type corresponding to the symptom training information to obtain a symptom training vector and a theoretical syndrome type vector corresponding to each symptom training information; the symptom training information is obtained by preprocessing the symptom description and tongue and pulse condition in the TCM syndrome differentiation theory and clinical cases; The syndrome differentiation module inputs the symptom training vector and the theoretical syndrome type vector into the first deep learning network to obtain the measured syndrome type corresponding to the symptom training information; The syndrome differentiation module extracts vectors from the measured syndrome types to obtain measured syndrome type vectors; The syndrome differentiation module iteratively updates the parameters and hyperparameters of the first deep learning network according to the loss information of the theoretical syndrome type vector and the measured syndrome type vector to obtain the deep learning syndrome differentiation model.

5. The Chinese medicine acupuncture decision-making assistance method according to claim 1, characterized in that: The prescription generating module generates an acupuncture prescription according to the syndrome type and the symptom information, including: The prescription generation module pre-processes the symptom information to obtain standard symptom data; The prescription generation module performs feature extraction on the standard symptom data and the syndrome type to obtain a symptom vector and a syndrome type vector; The prescription generation module inputs the symptom vector, the syndrome type vector and the TCM knowledge base into a prescription generation model to obtain an acupuncture prescription for the patient; the TCM knowledge base includes acupuncture treatment rules and acupoint combinations.

6. The Chinese medicine acupuncture decision-making assistance method according to claim 5, characterized in that: The prescription generation model is trained in the following way: The prescription generation module extracts features from each symptom training information, the syndrome type corresponding to the symptom training information, and the acupuncture point information corresponding to the symptom training information, to obtain a symptom training vector, a theoretical syndrome type vector, and an acupuncture point vector corresponding to each symptom training information; the symptom training information is obtained by pre-processing the symptom description and tongue and pulse images in the TCM syndrome differentiation theory and clinical cases; The prescription generation module inputs the symptom training vector, the theoretical syndrome vector and the acupuncture point vector into a second deep learning network to obtain a training prescription corresponding to the symptom training information; The prescription generation module extracts a vector from the training prescription to obtain a training prescription vector; The prescription generation module iteratively updates the parameters and hyperparameters of the second deep learning network according to the acupuncture point vector and the loss information of the training prescription to obtain the prescription generation model.

7. The Chinese medicine acupuncture decision-making assistance method according to claim 1, characterized in that: The three-dimensional human body model is constructed in the following manner: The acupoint display module obtains the patient's human anatomy data and acupoint data; The acupuncture point display module inputs the human anatomical data into a three-dimensional modeling tool to generate a three-dimensional original model of the patient; The acupoint display module adds the acupoint data to the three-dimensional original model and renders the three-dimensional original model to obtain the three-dimensional human body model of the patient.

8. The Chinese medicine acupuncture decision-making assistance method according to claim 1, characterized in that: The method further comprises: The user interaction interface transmits the prescription modification information to the prescription generation module; The prescription generating module updates the acupuncture prescription according to the prescription modification information to obtain an updated acupuncture prescription; The acupoint display module re-marks the three-dimensional human body model of the patient according to the updated acupuncture prescription.

9. A Chinese medicine acupuncture decision-making support system, characterized in that: The TCM acupuncture decision-making support system includes a collection module, a syndrome differentiation module, a prescription generation module, an acupoint display module and a user interaction interface; The acquisition module is used to transmit the patient's symptom information to the syndrome differentiation module; the symptom information includes the disease name and symptom information; The syndrome differentiation module is used to analyze the disease information using a deep learning syndrome differentiation model to obtain the syndrome type of the patient; the deep learning syndrome differentiation model is obtained based on TCM syndrome differentiation theory and clinical case training; The prescription generation module is used to generate an acupuncture prescription according to the syndrome type and the symptom information; the acupuncture prescription includes acupuncture point combination, acupoint location, acupuncture depth and acupuncture direction; The acupuncture point display module is used to mark the acupuncture point combination, the acupuncture point position, the acupuncture depth and the acupuncture direction on the three-dimensional human body model of the patient; the patient corresponds to the three-dimensional human body model one by one; The user interaction interface is used to display the syndrome type, the acupuncture prescription and the three-dimensional human body model so that the clinician can perform acupuncture treatment on the patient.

10. The Chinese medicine acupuncture decision-making support system according to claim 9, characterized in that: The syndrome differentiation module is used to pre-process the symptom information to obtain standard symptom data; The syndrome differentiation module is used to extract features from the standard syndrome data to obtain a syndrome vector; The syndrome differentiation module is used to input the symptom vector into the deep learning syndrome differentiation model, perform syndrome differentiation analysis, and obtain the patient's syndrome type.

Images