Human experience management system for traditional Chinese medicine based on artificial intelligence

Abstract

The invention relates to a traditional Chinese medicine personal experience management system based on artificial intelligence, and relates to the technical field of traditional Chinese medicine data analysis, the system comprises a block chain network composed of a public chain and a permission chain and a plurality of edge computing nodes, and the edge computing nodes are deployed in each medical institution; the edge computing node is used for collecting traditional Chinese medicine clinical data of the user, and the traditional Chinese medicine clinical data comprises traditional Chinese medicine prescription data and curative effect evaluation data; encrypting the traditional Chinese medicine clinical data based on a homomorphic encryption technology, and performing feature extraction on the encrypted traditional Chinese medicine clinical data based on an artificial intelligence technology; inputting the extracted feature information into a pre-constructed traditional Chinese medicine analysis model to obtain traditional Chinese medicine knowledge rule information; and the public chain is used for storing the traditional Chinese medicine knowledge rule information sent by each edge computing node, and determining multi-person experience information based on each piece of traditional Chinese medicine knowledge rule information. According to the invention, effective management and sharing of traditional Chinese medicine personal experience are realized, and the security and privacy of data are guaranteed.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine data analysis technology, and in particular to an artificial intelligence-based management system for human experience in using traditional Chinese medicine. Background Technology

[0002] Traditional Chinese medicine (TCM) prescriptions generated in clinical practice need to be summarized through "human experience" for application as new TCM drugs. "Human experience" refers to a general summary of TCM clinical diagnosis and treatment understanding accumulated over long-term clinical practice, which meets clinical needs and possesses certain regularity and reproducibility. Collecting and summarizing human experience data to form high-quality data and evaluable evidence is a crucial step in TCM "human experience" research.

[0003] In existing technologies, a drug or prescription usage database is used to extract prescription information and review it. Then, the patient's basic information, drug information, and relevant medication information are compared with the relevant information and rules for rational drug use guidance in the drug usage database to obtain and output medication guidance information. However, this technical solution is often applied within a closed hospital system. Medication information for the same patient may be recorded differently in different hospitals and cannot be shared. Furthermore, data on the use of the same traditional Chinese medicine prescription by multiple patients with similar symptoms is difficult to share and summarize, making it difficult to form valuable experience data. Moreover, data aggregation from multiple patients, multiple symptoms, or multiple prescriptions is often disorganized, making it difficult to form effective and valuable experience data. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an artificial intelligence-based human experience management system for traditional Chinese medicine, which aims to solve at least one of the above-mentioned technical problems.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: In the first aspect, this application provides an artificial intelligence-based management system for the human experience of using traditional Chinese medicine, which adopts the following technical solution: an artificial intelligence-based management system for the human experience of using traditional Chinese medicine, including a blockchain network composed of a public chain and a permissioned chain and multiple edge computing nodes, wherein the edge computing nodes are deployed in various medical institutions; The edge computing node is used to collect users' TCM clinical data, which includes user profiles, TCM prescription data, and efficacy evaluation data. The TCM clinical data is encrypted using homomorphic encryption technology to obtain encrypted TCM clinical data. Features are extracted from the encrypted TCM clinical data using artificial intelligence technology to obtain feature information. This feature information is then input into a pre-built TCM analysis model to obtain the user's TCM knowledge rule information, which includes the applicable disease range, optimal dosage range, contraindications, and efficacy influencing factors of the TCM. The public blockchain is used to store the traditional Chinese medicine knowledge rule information sent by each edge computing node, and to determine multiple user experience information based on each of the traditional Chinese medicine knowledge rule information, and to construct a traditional Chinese medicine knowledge graph based on each of the user experience information. The permissioned chain is used to manage permissions for account information and nodes in the blockchain network. The account information includes doctors and regulators, and the permission information includes data access permissions, data modification permissions, and model training permissions.

[0006] The beneficial effects of this invention are: it enables the collection and processing of users' clinical data on traditional Chinese medicine using edge computing nodes, and after encryption and feature extraction, inputs the data into a traditional Chinese medicine analysis model to obtain users' knowledge rules information on traditional Chinese medicine; the public blockchain can store this information and construct a knowledge graph of traditional Chinese medicine; the permissioned blockchain can manage the permissions of accounts and nodes in the blockchain network, realizing the effective management and sharing of experience in using traditional Chinese medicine, improving the efficiency and accuracy of clinical research and application of traditional Chinese medicine, and ensuring the security and privacy of data.

[0007] Based on the above technical solution, the present invention can be further improved as follows.

[0008] Furthermore, the edge computing node includes a medication dispensing module, which is specifically used for: Acquire user symptom information, symptom images, and user profiles, wherein the symptom information is either voice information or text information; The user's symptom information is segmented into words to obtain multiple keywords related to the symptom information; All the keywords are classified based on a preset text classification model to obtain the dimension to which each keyword belongs, and the dimension represents the domain of the corresponding keyword; Based on a preset semantic analysis technique, semantic analysis is performed on all the keywords to obtain a semantic description of each keyword, and the semantic description represents the medical terminology of the keyword; Based on a pre-set sensitive word library, the sensitivity level of each keyword is determined. The pre-set sensitive word library is set based on medical contraindications, social media public opinion hotspots, and public health emergency events. The symptom images are used to identify features based on image recognition technology to obtain symptom features; The semantic description and symptom features of each keyword are fused together, and a multimodal fusion method based on attention mechanism is used to obtain the symptom description text. Based on the user's corresponding symptom description text, the sensitivity of multiple keywords, and a pre-constructed traditional Chinese medicine knowledge graph, multiple initial prescription plans for the user are determined; Based on the user's historical medication information, user profile, and preset medication evaluation rules, each initial medication plan is evaluated to obtain an evaluation score for each initial medication plan; Each of the initial medication dispensing protocols and the evaluation score for each initial medication dispensing protocol are sent to the doctor's terminal so that the doctor can dispense medication for the user based on the evaluation scores.

[0009] The beneficial effects of adopting the above-mentioned further scheme are: it can collect users' symptom information, symptom images, and user profiles, and process them through word segmentation, classification, semantic analysis, sensitivity determination, and image feature recognition to obtain symptom description text. Combining the symptom description text, keyword sensitivity, and traditional Chinese medicine knowledge graph, multiple initial prescription plans are determined. Then, based on the user's historical prescription information, user profile, and prescription evaluation rules, the initial prescription plans are evaluated to obtain evaluation scores. Finally, the plans and scores are sent to doctors for reference, which helps doctors to prescribe medication for users more accurately and efficiently.

[0010] Furthermore, the medication dispensing module, when evaluating each initial medication dispensing plan based on the user's historical medication dispensing information, user profile, and preset medication dispensing evaluation rules, and obtaining an evaluation score for each initial medication dispensing plan, specifically uses the following: Obtain historical evaluation information for each of the initial drug preparation schemes; Based on the historical evaluation information of each initial medication regimen, a first weight value is determined for each initial medication regimen; Based on the user's symptom information, symptom images, user profile, and each of the initial medication regimens, the expected efficacy of each initial medication regimen is evaluated in the artificial intelligence model. Based on the expected efficacy of each of the initial medication regimens, a second weight value is determined for each of the initial medication regimens; Based on the degree of synergistic and antagonistic influence between each Chinese herb in each initial dispensing scheme, a third weight value is determined for each initial dispensing scheme; Based on the user's medical history, age, gender, allergy history, each initial medication regimen, and a preset neural network model in the user profile, the probability of adverse reactions occurring when the user uses each initial medication regimen is predicted; Based on the probability of adverse reactions for each initial medication regimen, a fourth weight value is determined for each initial medication regimen; A fifth weight value is determined for each initial medication regimen based on its price, efficacy, and the user's preferences in the user profile. Obtain the inventory information of each Chinese medicine corresponding to each initial dispensing plan from each medical institution, and determine the sixth weight value of each initial dispensing plan based on the inventory information of each Chinese medicine corresponding to each initial dispensing plan. An evaluation score is determined for each initial medication preparation scheme based on the first, second, third, fourth, fifth, and sixth weight values.

[0011] The beneficial effects of adopting the above-mentioned further approach are as follows: By acquiring historical evaluation information for each initial medication regimen and determining the first weight value accordingly, past patient feedback on the regimen can be considered. Using an artificial intelligence model, combined with user symptom information, symptom images, and user profiles, the expected efficacy of each initial medication regimen is evaluated, and a second weight value is determined, which helps predict the actual effect of the regimen on the user. Analyzing the synergistic and antagonistic effects between each herb in each initial medication regimen determines the third weight value, which helps avoid adverse drug reactions. Based on user profiles and a pre-set neural network model, the probability of adverse reactions occurring when using each initial medication regimen is predicted, and a fourth weight value is determined accordingly, which helps identify potential risks in advance. Combining the price, efficacy, and user preferences in the user profile of each initial medication regimen, a fifth weight value is determined, which helps improve user satisfaction by considering the user's economic affordability and personal preferences while ensuring efficacy. Acquiring inventory information of each herb corresponding to each initial medication regimen from various medical institutions and determining the sixth weight value accordingly. This helps ensure the selected treatment plan is feasible in practice, avoiding medication delays or failures due to insufficient inventory, and improving the medication experience. It also helps medical personnel quickly select the optimal plan from among many options, improving work efficiency.

[0012] Furthermore, the edge computing node also includes a traditional Chinese medicine clinical data acquisition module, which is specifically used for: After the doctor dispenses the medication to the user, the corresponding initial medication template is obtained based on the final medication information. The medication template includes medical institution information, patient basic information, disease diagnosis information, medication information, medication guidance information, and multiple time content nodes. The medication information includes the name, dosage form, specifications, and quantity of the Chinese medicine. The time content nodes are used to record medication feedback information later. Based on the user profile, the initial medication template is adjusted to obtain a new medication template. The user profile includes lifestyle habits, past medical history, and allergy history. The new medication template is sent to the user's terminal so that the user can fill in medication feedback information in multiple time content nodes corresponding to the new medication template according to the actual medication situation, and generate a target medication template. The medication feedback information includes medication method and dosage, medication time, symptom pictures, description or consultation voice or text information, and one or more other medication information. After the user fills in the medication feedback information, the validity of the target medication template is verified. The validity verification includes data integrity verification, data rationality verification, and data consistency verification. If the validity of the user's target medication template is verified, the target medication template is preprocessed to obtain the user's traditional Chinese medicine clinical data.

[0013] The beneficial effects of adopting the above-mentioned further solutions are as follows: Using the TCM clinical data collection module to obtain initial medication templates after doctors prescribe medication, and adjusting the templates based on user profiles, makes the templates more closely match the actual situation of users; allowing users to fill in medication feedback information at the new template time points allows for the collection of actual medication information; validating the target medication templates ensures data quality; preprocessing valid templates yields accurate user TCM clinical data, and combining this with the existing functions of the blockchain network and edge computing nodes improves the data collection process of the TCM user experience management system.

[0014] Furthermore, each of the aforementioned edge computing nodes, when constructing a traditional Chinese medicine analysis model, is specifically used for: Obtain a training set, which includes multiple homomorphically encrypted historical clinical experience information of traditional Chinese medicine. The historical clinical experience information of traditional Chinese medicine consists of changes in symptoms and signs of patients after medication, as well as time-series characteristics of drug efficacy, arranged in a time sequence. Based on a bidirectional LSTM network, the patient's symptom changes and signs after medication are arranged in time series as input, and the drug efficacy time series features are output. Based on graph convolutional neural networks, the traditional Chinese medicine knowledge graph is used as input and prescription compatibility features are used as output; Based on the outputs of the time-series pharmacodynamics channel and the compatibility analysis channel, the feature fusion weights of the two channels are dynamically adjusted through a gating mechanism, and the fused features are output. The patient's syndrome type, current season, and regional information are encoded to generate a dynamic adjustment coefficient matrix, which is used to dynamically adjust the weight parameters in the dual-channel neural network model. Initialize the dual-channel neural network model according to the training set, calculate the encryption gradient, and send the encryption gradient to the public chain; The public blockchain is used to obtain the encryption gradient of each edge node, perform a weighted average based on the dynamic aggregation weight to obtain the aggregated encryption gradient, and send the aggregated encryption gradient to each edge computing node. Each of the aforementioned edge computing nodes is also used to update the dual-channel neural network model based on the aggregated encrypted gradient, thereby constructing a traditional Chinese medicine analysis model.

[0015] The beneficial effects of adopting the above-mentioned further scheme are as follows: A bidirectional LSTM network can effectively mine the temporal features of drug efficacy in patient symptom changes and physical signs data after medication; a graph convolutional neural network can output prescription compatibility features from a traditional Chinese medicine knowledge graph. Constructing a dual-channel neural network model can comprehensively consider both temporal drug efficacy and compatibility analysis, and the dynamic interaction gate can flexibly adjust feature fusion weights, improving the feature fusion effect. Furthermore, the model weight parameters can be dynamically adjusted based on patient syndrome differentiation, season, and regional information to adapt to drug efficacy change analysis under different conditions. Federated learning iterative training protects data privacy while improving model performance by combining multi-node data; dynamic weight aggregation balances data quality and node authority. Dynamic model updates employ incremental learning algorithms, enabling the model to adapt to new clinical feedback data in a timely manner, maintaining the model's accuracy and adaptability. Simultaneously, homomorphic encryption technology is used throughout the process to ensure data privacy and security.

[0016] Furthermore, the public blockchain, when determining multiple users' experience information based on the various traditional Chinese medicine knowledge rules, is specifically used for: According to the categories of Chinese medicine and the types of diseases, the information on the rules of Chinese medicine knowledge is classified to obtain multiple Chinese medicine knowledge datasets; For any of the traditional Chinese medicine knowledge datasets, determine whether there are different traditional Chinese medicine knowledge rule information obtained by the same user on multiple edge computing nodes in the traditional Chinese medicine knowledge datasets; For any of the traditional Chinese medicine knowledge datasets, if there are different traditional Chinese medicine knowledge rule information obtained by the same user on multiple edge computing nodes, then the traditional Chinese medicine knowledge rules of the same patient on different edge computing nodes are associated and integrated to obtain new traditional Chinese medicine knowledge rule information. For any given set of traditional Chinese medicine knowledge datasets, multiple sets of human experience information on traditional Chinese medicine are obtained based on each traditional Chinese medicine knowledge rule in the dataset.

[0017] The beneficial effects of adopting the above-mentioned further solutions are as follows: By classifying TCM knowledge rules information according to the categories of TCM and disease types, a structured TCM knowledge dataset can be formed. This helps to systematically organize scattered and fragmented TCM knowledge, improving the manageability and accessibility of information. Judging and integrating different TCM knowledge rules information obtained by the same user from multiple edge computing nodes can eliminate information silos and form a more complete and accurate user medication record. This helps to gain a more comprehensive understanding of the user's medication history and effects, providing stronger support for subsequent prescription and medication guidance. Storing and sharing TCM knowledge rules information and user experience information through a public blockchain can form a cross-institutional and cross-regional TCM knowledge sharing repository, helping to break down information barriers.

[0018] Furthermore, the public blockchain is also used for: When more than two pieces of human experience information are detected for the same traditional Chinese medicine knowledge point, the smart contract is triggered, and the human experience information for the traditional Chinese medicine knowledge point is updated according to the voting results of the smart contract. The smart contract is used to organize relevant medical institution nodes to vote on the same traditional Chinese medicine knowledge point using experience information from multiple people, obtain voting information, and determine the voting result based on each voting information and a preset voting weight calculation rule.

[0019] The beneficial effects of adopting the above-mentioned further solution are: when the public blockchain detects that there are more than two pieces of human experience information on the same traditional Chinese medicine knowledge point, it can trigger a smart contract, allowing relevant medical institution nodes to vote on these pieces of human experience information, and update the human experience information on the traditional Chinese medicine knowledge point according to the voting results, thereby achieving effective management and updating of human experience information and ensuring the accuracy and reliability of traditional Chinese medicine knowledge.

[0020] Secondly, this application provides a method for managing human experience in traditional Chinese medicine based on artificial intelligence, employing the following technical solution: An artificial intelligence-based method for managing human experience in traditional Chinese medicine includes: The system collects users' traditional Chinese medicine (TCM) clinical data, which includes user profiles, TCM prescription data, and efficacy evaluation data. The TCM clinical data is then encrypted using homomorphic encryption technology to obtain encrypted TCM clinical data. Feature information is obtained by extracting features from the encrypted clinical data of traditional Chinese medicine based on artificial intelligence technology. The feature information is input into a pre-built traditional Chinese medicine analysis model to obtain the user's traditional Chinese medicine knowledge rule information, which includes the applicable disease range, optimal dosage range, contraindication information, and efficacy influencing factors of traditional Chinese medicine. The traditional Chinese medicine knowledge rules information is sent to a blockchain network for storage, so that the blockchain network can determine multiple user experience information based on the traditional Chinese medicine knowledge rules information sent by each edge computing node, and construct a traditional Chinese medicine knowledge graph based on each user experience information.

[0021] Thirdly, this application provides an electronic device that adopts the following technical solution: An electronic device includes a memory and a processor, wherein the memory stores a computer program capable of being loaded by the processor and executing the artificial intelligence-based human experience management method for traditional Chinese medicine as described in the second aspect.

[0022] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the artificial intelligence-based human experience management method for traditional Chinese medicine as described in the second aspect.

[0023] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0024] Figure 1 A schematic diagram of the structure of an artificial intelligence-based human experience management system for traditional Chinese medicine provided in one embodiment of the present invention; Figure 2 A flowchart illustrating an artificial intelligence-based method for managing human experience in traditional Chinese medicine, provided as an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device according to the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0027] like Figure 1 As shown, this application embodiment provides an artificial intelligence-based TCM human experience management system 100. The system includes a blockchain network composed of a public chain 102 and a permissioned chain 103 and multiple edge computing nodes 101, which are deployed in various medical institutions. Each edge computing node 101 is used to collect users' traditional Chinese medicine (TCM) clinical data, including user profiles, TCM prescription data, and efficacy evaluation data. The TCM clinical data is encrypted using homomorphic encryption technology to obtain encrypted TCM clinical data. Features are extracted from the encrypted TCM clinical data using artificial intelligence technology to obtain feature information. This feature information is then input into a pre-built TCM analysis model to obtain the user's TCM knowledge rule information, which includes the applicable disease range, optimal dosage range, contraindications, and efficacy influencing factors of the TCM. The public chain 103 is used to store the traditional Chinese medicine knowledge rule information sent by each edge computing node 101, and to determine multiple human experience information based on each of the traditional Chinese medicine knowledge rule information, and to construct a traditional Chinese medicine knowledge graph based on each of the human experience information. The permission chain 103 is used to manage the permissions of account information and nodes in the blockchain network. The account information includes doctors and regulators, and the permission information includes data access permissions, data modification permissions, and model training permissions.

[0028] In this embodiment, a blockchain is essentially a ledger that maintains public data. It can be understood as a specially designed database or a distributed database. Its decentralized or multi-centralized approach enables consensus algorithms to achieve consistency of ledger data among blockchain nodes. Public chain 102 is a fully decentralized blockchain, while permissioned chain 103 is a partially decentralized blockchain whose ledger data is open to specific users or organizations, requiring authorization to participate in the blockchain's maintenance, data reading and writing, and consensus processes.

[0029] In this embodiment, after edge computing node 101 encrypts the original clinical data using homomorphic encryption technology, edge computing node 101 generates a first data fingerprint for the encrypted original clinical data using a hash algorithm. Edge computing node 101 stores the encrypted original clinical data in its local secure storage area and simultaneously sends a data storage request to permissioned blockchain 103. The request includes the first data fingerprint and basic data information. Upon receiving the request from edge computing node 101, permissioned blockchain 103 verifies the request, checking whether the data format meets requirements and whether the data source is legal. After successful verification, permissioned blockchain 103 stores the first data fingerprint in its own blockchain ledger and sends the successfully stored first data fingerprint back to edge computing node 101. Edge computing node 101 records this feedback information as proof that the data has been stored on the blockchain.

[0030] In this embodiment of the application, each edge computing node 101 includes a drug dispensing module, which is specifically used to perform the following steps: S11, Obtain the user's symptom information, symptom images, and user profile, wherein the symptom information is voice information or text information; In this embodiment of the application, the symptom information is the user's description of their own symptoms, such as: "I have been experiencing headaches and nausea recently, especially when I wake up in the morning." The symptom image can be an image of the user's skin lesions, etc.

[0031] S12, perform word segmentation on the user's symptom information to obtain multiple keywords of the symptom information; In this embodiment of the application, word segmentation tools in natural language processing are used to segment the symptom information. Taking "I have been feeling headaches recently, and also have nausea, especially when I get up in the morning" as an example, the word segmentation result may be "I / recently / feel / headache / , / also / with / nausea / , / especially / when / I / get / up / in the morning / particularly / obvious".

[0032] S13, classify all the keywords based on a preset text classification model to obtain the dimension to which each keyword belongs, wherein the dimension represents the domain of the corresponding keyword; In this embodiment, the pre-defined text classification model can be trained using a deep learning model. During model training, a large amount of labeled symptom keyword data is used to classify the keywords into different dimensions, such as symptom dimension, time dimension, and severity dimension. During the classification process, the keywords obtained from word segmentation are input into the trained model to obtain the dimension to which each keyword belongs.

[0033] S14, Based on a preset semantic analysis technique, perform semantic analysis on all the keywords to obtain a semantic description of each keyword, wherein the semantic description represents the medical terminology of the keyword; S15, Based on a preset sensitive word library, determine the sensitivity level of each keyword. The preset sensitive word library is set based on medical contraindications, social media public opinion hotspots, and public health emergency events. In this embodiment, the pre-defined sensitive word library includes sensitive words set based on medical contraindications, social media trending topics, and public health emergencies. For example, regarding medical contraindications, "penicillin" is a sensitive word for users with a history of penicillin allergy; regarding social media trending topics, if a certain traditional Chinese medicine is recently exposed to have quality problems, its name will also be included in the sensitive word library; regarding public health emergencies, for example, during an epidemic, symptoms and drugs related to the epidemic may be set as sensitive words. When determining the sensitivity of keywords, the keywords are matched with the sensitive word library, and the sensitivity is determined based on the matching results.

[0034] S16, Based on image recognition technology, feature recognition is performed on the symptom image to obtain symptom features; In this embodiment, if the user inputs a symptom image, image recognition technology, such as a convolutional neural network (CNN) model, is used to identify features of the symptom image. Taking a skin lesion image as an example, the model can identify features such as the color, shape, size, and boundary of the lesion.

[0035] S17, The semantic description and symptom features of each keyword are fused together, and a multimodal fusion method based on attention mechanism is used to obtain the symptom description text; In this embodiment of the application, if the user inputs a symptom image, a multimodal fusion method based on an attention mechanism is used to fuse the semantic description of the keywords and the symptom features.

[0036] S18. Based on the symptom description text corresponding to the user, the sensitivity of multiple keywords, and the pre-constructed traditional Chinese medicine knowledge graph, determine multiple initial prescription plans for the user. In this embodiment, the pre-constructed traditional Chinese medicine (TCM) knowledge graph contains information such as the properties (e.g., flavor, meridian tropism, efficacy), applicable symptoms, and compatibility relationships of TCM herbs. When determining the initial dispensing plan, the dispensing module queries and infers from the TCM knowledge graph based on the symptom description text. For example, based on the symptom of "headache," it queries for TCM herbs with analgesic effects, such as Chuanxiong (Ligusticum striatum) and Baizhi (Angelica dahurica).

[0037] Simultaneously, the sensitivity of keywords is considered, and highly sensitive traditional Chinese medicines are excluded. For example, if "a certain traditional Chinese medicine" is identified as a highly sensitive word, it will be excluded when generating the prescription. Finally, based on the symptom description text and the traditional Chinese medicine knowledge graph, multiple initial prescription schemes are determined. For example, Scheme 1 is Chuanxiong, Baizhi, and Tianma; Scheme 2 is Chuanxiong, Juhua, and Jue Mingzi, etc.

[0038] S19, based on the user's historical medication information, user profile and preset medication evaluation rules, evaluate each initial medication plan to obtain an evaluation score for each initial medication plan; In this embodiment of the application, the medication dispensing module is used to evaluate each initial medication dispensing plan based on the user's historical medication dispensing information, user profile, and preset medication dispensing evaluation rules, and to obtain an evaluation score for each initial medication dispensing plan, specifically for: Obtain historical evaluation information for each of the initial drug preparation schemes; Based on the historical evaluation information of each initial medication regimen, a first weight value is determined for each initial medication regimen; Based on the user's symptom information, symptom images, user profile, and each of the initial medication regimens, the expected efficacy of each initial medication regimen is evaluated in the artificial intelligence model. Based on the expected efficacy of each of the initial medication regimens, a second weight value is determined for each of the initial medication regimens; Based on the degree of synergistic and antagonistic influence between each Chinese herb in each initial dispensing scheme, a third weight value is determined for each initial dispensing scheme; Based on the user's medical history, age, gender, allergy history, each initial medication regimen, and a preset neural network model in the user profile, the probability of adverse reactions occurring when the user uses each initial medication regimen is predicted; Based on the probability of adverse reactions for each initial medication regimen, a fourth weight value is determined for each initial medication regimen; A fifth weight value is determined for each initial medication regimen based on its price, efficacy, and the user's preferences in the user profile. Obtain the inventory information of each Chinese medicine corresponding to each initial dispensing plan from each medical institution, and determine the sixth weight value of each initial dispensing plan based on the inventory information of each Chinese medicine corresponding to each initial dispensing plan. An evaluation score is determined for each initial medication preparation scheme based on the first, second, third, fourth, fifth, and sixth weight values.

[0039] In this embodiment, historical evaluation information for each initial medication regimen is collected from multiple channels, including electronic medical record systems of medical institutions, patient feedback platforms, and online medical communities. This historical evaluation information is then quantitatively analyzed, and the frequency of occurrence of each initial medication regimen at different evaluation levels is counted. The average score is calculated to obtain the first weight value for each initial medication regimen. Based on a traditional Chinese medicine pharmacology knowledge base, the synergistic and antagonistic effects between each traditional Chinese medicine in each initial medication regimen are analyzed to determine the third weight value for each initial medication regimen.

[0040] In this embodiment, a pre-defined neural network model is constructed, using the user's medical history (such as hypertension, diabetes, etc.), age, gender, allergy history, and each initial medication regimen from the user profile as input, and the probability of adverse reactions occurring after the patient uses the regimen as output for training. Training data can come from a medical institution's adverse reaction report database. The current user's information and each initial medication regimen are input into the trained neural network model to predict the probability of adverse reactions occurring when the user uses each initial medication regimen. This determines the fourth weight value.

[0041] In this embodiment, price information for each initial dispensing plan can be obtained from a drug procurement system or a drug price database. The user's preferences are analyzed, such as whether the user prioritizes affordability, efficacy, or the naturalness of traditional Chinese medicine. Taking into account price, efficacy, and user preferences, a weighted summation method is used to determine the fifth weight value.

[0042] In this embodiment, the system interfaces with the drug inventory management systems of various medical institutions to obtain the inventory information of each traditional Chinese medicine corresponding to each initial dispensing plan. Based on the inventory information, a sixth weight value is calculated for each initial dispensing plan.

[0043] Finally, assign reasonable weight coefficients to each weight value. For example, the weight coefficient for the first weight value is 0.1, the second weight value is 0.3, the third weight value is 0.2, the fourth weight value is 0.2, the fifth weight value is 0.1, and the sixth weight value is 0.1. These weight coefficients can be adjusted according to actual needs and expert experience. A weighted summation method is used to calculate the evaluation score for each initial medication regimen.

[0044] S20, each of the initial medication dispensing plans and the evaluation score of each initial medication dispensing plan are sent to the doctor's terminal so that the doctor can dispense medication for the user with reference to the evaluation scores.

[0045] In this embodiment, each initial medication dispensing plan and its evaluation score are encapsulated into a standardized data format. For example, JSON format is used, and the data includes information such as the plan name, Chinese herbal medicine composition, and evaluation score. The encapsulated data is sent to the doctor's terminal via a network communication protocol (such as HTTP, WebSocket, etc.). The doctor's terminal can be a computer, tablet, or mobile phone. After receiving the data on the terminal, the doctor can view each initial medication dispensing plan and its evaluation score, and, based on their clinical experience and professional knowledge, refer to the evaluation score to dispense medication for the user. For example, the doctor may choose Plan One, which has a higher evaluation score, and make appropriate adjustments according to the patient's specific condition.

[0046] By acquiring users' symptom information, symptom images, and user profiles, and processing them through word segmentation, classification, semantic analysis, sensitivity determination, and image feature recognition, symptom description text is obtained. Combining the symptom description text, keyword sensitivity, and traditional Chinese medicine knowledge graph, multiple initial prescription plans are determined. Then, based on the user's historical prescription information, user profile, and prescription evaluation rules, the initial prescription plans are evaluated to obtain an evaluation score. Finally, the plans and scores are sent to doctors for reference, which helps doctors to prescribe medication for users more accurately and efficiently.

[0047] Optionally, each of the edge computing nodes 101 further includes a traditional Chinese medicine clinical data acquisition module, which is specifically used for: After the doctor dispenses the medication to the user, the corresponding initial medication template is obtained based on the final medication information. The medication template includes medical institution information, patient basic information, disease diagnosis information, medication information, medication guidance information, and multiple time content nodes. The medication information includes the name, dosage form, specifications, and quantity of the Chinese medicine. The time content nodes are used to record medication feedback information later. Based on the user profile, the initial medication template is adjusted to obtain a new medication template. The user profile includes lifestyle habits, past medical history, and allergy history. The new medication template is sent to the user's terminal so that the user can fill in medication feedback information in multiple time content nodes corresponding to the new medication template according to the actual medication situation, and generate a target medication template. The medication feedback information includes medication method and dosage, medication time, symptom pictures, description or consultation voice or text information, and one or more other medication information. After the user fills in the medication feedback information, the validity of the target medication template is verified. The validity verification includes data integrity verification, data rationality verification, and data consistency verification. If the validity of the user's target medication template is verified, the target medication template is preprocessed to obtain the user's traditional Chinese medicine clinical data.

[0048] In this embodiment, after the doctor completes the medication dispensing process for the user, the traditional Chinese medicine clinical data acquisition module of the edge computing node 101 interacts with the medical institution's medication dispensing system. It obtains the final medication information corresponding to this dispensing from the system and then matches the corresponding initial medication template from the database according to preset rules. For example, the database stores various initial medication templates according to dimensions such as medical institution and disease type. When medication information for a cold from a tertiary hospital is obtained, the system finds the corresponding initial medication template for the cold from that hospital. Afterwards, the initial medication template is adjusted based on the user profile. For example, considering that prolonged sleep deprivation may affect the user's recovery, a prompt "It is recommended to ensure sufficient sleep and avoid staying up late" is added to the medication guidance information. Then, the user fills in medication feedback information in multiple time-related content nodes corresponding to the new medication template based on the actual medication situation.

[0049] After the user fills in the medication feedback information, the system checks whether all necessary information has been entered into the target medication template. For example, it checks whether medication feedback information has been entered for each time point, and whether key information such as dosage and duration of medication is complete. If any information is missing, the system prompts the user to supplement it. The system then assesses the reasonableness of the user-entered data. For example, it checks whether the dosage is within the normal range, whether the duration of medication conforms to the drug's instructions, and whether the symptom images match the described symptoms. If unreasonable data is found, the system marks it. The system also checks the consistency of data between different time points. For example, if the user describes worsening symptoms at one time point and complete symptom disappearance at another, but the dosage and duration of medication are not adjusted accordingly, the system considers the data potentially inconsistent and requires further confirmation. If the user's target medication template passes the validity verification, it undergoes preprocessing. Preprocessing includes data cleaning, data transformation, and data standardization, ultimately yielding the user's traditional Chinese medicine clinical data.

[0050] In this embodiment of the application, the edge computing node 101 is used to encrypt the user's traditional Chinese medicine clinical data based on homomorphic encryption technology. When the encrypted traditional Chinese medicine clinical data is obtained, it is specifically used for: The clinical data of traditional Chinese medicine is encrypted using a symmetric key to obtain a first ciphertext of clinical data of traditional Chinese medicine; the first ciphertext of clinical data of traditional Chinese medicine is then encrypted using a public key to obtain a second ciphertext of clinical data of traditional Chinese medicine; the ciphertext of clinical data of traditional Chinese medicine is then sent to a server, so that the server generates encrypted clinical data of traditional Chinese medicine based on the second ciphertext of clinical data of traditional Chinese medicine, the second ciphertext of clinical data of traditional Chinese medicine sent through the data center and a preset homomorphic encryption technology, and obtains the encrypted clinical data of traditional Chinese medicine sent by the server. The second ciphertext of clinical data of traditional Chinese medicine is clinical data of traditional Chinese medicine that has been encrypted twice.

[0051] In this embodiment of the application, each edge computing node 101 is used to construct a traditional Chinese medicine analysis model, specifically for: obtaining a training set, the training set including multiple homomorphically encrypted historical traditional Chinese medicine clinical experience information, the historical traditional Chinese medicine clinical experience information being patient symptom changes, signs data and drug efficacy time sequence characteristics arranged in time series; Based on a bidirectional LSTM network, the patient's symptom changes and signs after medication are arranged in time series as input, and the drug efficacy time series features are output. Based on graph convolutional neural networks, the traditional Chinese medicine knowledge graph is used as input and prescription compatibility features are used as output; Based on the outputs of the time-series pharmacodynamics channel and the compatibility analysis channel, the feature fusion weights of the two channels are dynamically adjusted through a gating mechanism, and the fused features are output. The patient's syndrome type, current season, and regional information are encoded to generate a dynamic adjustment coefficient matrix, which is used to dynamically adjust the weight parameters in the dual-channel neural network model. Initialize the dual-channel neural network model according to the training set, calculate the encryption gradient, and send the encryption gradient to the public chain 102; The public chain 102 is used to obtain the encryption gradient of each edge node, perform a weighted average according to the dynamic aggregation weight to obtain the aggregated encryption gradient, and send the aggregated encryption gradient to each edge computing node 101. Each of the edge computing nodes 101 is also used to update the dual-channel neural network model based on the aggregated encrypted gradient to construct a traditional Chinese medicine analysis model.

[0052] In this embodiment of the application, the aggregation weight = α × the case data quality score of the node + β × the node authority coefficient; wherein, the data quality score is calculated from indicators such as the completeness of the data of the node, the proportion of follow-up data, and the standardization of case records; the node authority coefficient is determined by authoritative information such as hospital level and number of experts.

[0053] When edge computing node 101 receives new clinical feedback data, each edge node updates its local model using an incremental learning algorithm, calculates the loss increment of the new data on the current model, fine-tunes the model parameters based on the loss increment, submits the adjustment amount of the model parameters to the central server, the central server aggregates the adjustment amounts of each node in a federated averaging manner, updates the global model, and distributes the updated global model to each edge node.

[0054] By utilizing a bidirectional LSTM network, the temporal characteristics of drug efficacy in patient symptom changes and signs after medication can be effectively mined. A graph convolutional neural network can output prescription compatibility features from a traditional Chinese medicine knowledge graph. Constructing a dual-channel neural network model comprehensively considers both temporal efficacy and compatibility analysis, and the dynamic interaction gate can flexibly adjust feature fusion weights to improve feature fusion effectiveness. Furthermore, the model weight parameters can be dynamically adjusted based on patient syndrome differentiation, season, and regional information to adapt to efficacy change analysis under different conditions. Federated learning iterative training protects data privacy while improving model performance by combining multi-node data; dynamic weight aggregation balances data quality and node authority. Dynamic model updates employ incremental learning algorithms, enabling the model to adapt to new clinical feedback data in a timely manner, maintaining model accuracy and adaptability. Simultaneously, homomorphic encryption technology is used throughout the process to ensure data privacy and security.

[0055] Optionally, the public blockchain, when used to determine multiple users' experience information based on the various traditional Chinese medicine knowledge rules, is specifically used for: According to the categories of Chinese medicine and the types of diseases, the information on the rules of Chinese medicine knowledge is classified to obtain multiple Chinese medicine knowledge datasets; For any of the traditional Chinese medicine knowledge datasets, determine whether there are different traditional Chinese medicine knowledge rule information obtained by the same user on multiple edge computing nodes 101 in the traditional Chinese medicine knowledge datasets; For any of the traditional Chinese medicine knowledge datasets, if there are different traditional Chinese medicine knowledge rule information obtained by the same user on multiple edge computing nodes 101, then the traditional Chinese medicine knowledge rules of the same patient on different edge computing nodes 101 are associated and integrated to obtain new traditional Chinese medicine knowledge rule information. For any given set of traditional Chinese medicine knowledge datasets, multiple sets of human experience information on traditional Chinese medicine are obtained based on each traditional Chinese medicine knowledge rule in the dataset.

[0056] By classifying traditional Chinese medicine (TCM) knowledge rules according to the categories of TCM herbs and disease types, a structured TCM knowledge dataset can be formed. This helps to systematically organize scattered and fragmented TCM knowledge, improving information manageability and accessibility. Judging and integrating different TCM knowledge rule information obtained by the same user from multiple edge computing nodes 101 can eliminate information silos and form more complete and accurate user medication records. This helps to gain a more comprehensive understanding of the user's medication history and effects, providing stronger support for subsequent prescription and medication guidance. Storing and sharing TCM knowledge rule information and user experience information through a public blockchain can form a cross-institutional and cross-regional TCM knowledge sharing repository, helping to break down information barriers.

[0057] Optionally, public chain 102 can also be used for: When more than two pieces of human experience information are detected for the same traditional Chinese medicine knowledge point, the smart contract is triggered, and the human experience information for the traditional Chinese medicine knowledge point is updated according to the voting results of the smart contract. The smart contract is used to organize relevant medical institution nodes to vote on the same traditional Chinese medicine knowledge point using experience information from multiple people, obtain voting information, and determine the voting result based on each voting information and a preset voting weight calculation rule.

[0058] In this embodiment of the application, the weight is calculated based on the medical institution's historical valid data volume, total data volume, and institution authority level score.

[0059] This AI-based TCM (Traditional Chinese Medicine) user experience management system 100 utilizes edge computing nodes 101 to collect and process users' TCM clinical data. After encryption and feature extraction, the data is input into a TCM analysis model to obtain users' TCM knowledge rules information. A public blockchain 102 can store this information and construct a TCM knowledge graph. A permissioned blockchain 103 can manage the permissions of accounts and nodes on the blockchain network, achieving effective management and sharing of TCM user experience, improving the efficiency and accuracy of TCM clinical research and application, and ensuring data security and privacy.

[0060] Figure 2 A flowchart illustrating an artificial intelligence-based method for managing human experience in traditional Chinese medicine is shown.

[0061] like Figure 2 As shown, an artificial intelligence-based method for managing human experience in traditional Chinese medicine includes: The system collects users' traditional Chinese medicine (TCM) clinical data, which includes user profiles, TCM prescription data, and efficacy evaluation data. The TCM clinical data is then encrypted using homomorphic encryption technology to obtain encrypted TCM clinical data. Feature information is obtained by extracting features from the encrypted clinical data of traditional Chinese medicine based on artificial intelligence technology. The feature information is input into a pre-built traditional Chinese medicine analysis model to obtain the user's traditional Chinese medicine knowledge rule information, which includes the applicable disease range, optimal dosage range, contraindication information, and efficacy influencing factors of traditional Chinese medicine. The traditional Chinese medicine knowledge rules information is sent to a blockchain network for storage, so that the blockchain network can determine multiple user experience information based on the traditional Chinese medicine knowledge rules information sent by each edge computing node, and construct a traditional Chinese medicine knowledge graph based on each user experience information.

[0062] It should be noted that the execution subject of this method can be an electronic device, which can be a server or a mobile terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The mobile terminal device can be a laptop, desktop computer, etc., but is not limited to these. Its implementation principle is the same as that of the artificial intelligence-based TCM human experience management system described earlier, and will not be repeated here.

[0063] Figure 3 This is a structural block diagram of an electronic device 300 according to an embodiment of this application.

[0064] like Figure 3 As shown, the electronic device 300 includes a processor 301 and a memory 302, and may further include one or more of an information input / output (I / O) interface 303, a communication component 304, and a communication bus 305.

[0065] The processor 301 controls the overall operation of the electronic device 300 to complete all or part of the steps in the aforementioned artificial intelligence-based method for managing human experience in traditional Chinese medicine. The memory 302 stores various types of data to support the operation of the electronic device 300. This data may include, for example, instructions for any application or method operating on the electronic device 300, as well as application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0066] I / O interface 303 provides an interface between processor 301 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 304 is used to test wired or wireless communication between electronic device 300 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 304 may include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0067] The communication bus 305 may include a path for transmitting information between the aforementioned components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 can be divided into an address bus, a data bus, a control bus, etc.

[0068] The electronic device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the artificial intelligence-based traditional Chinese medicine human experience management method given in the above embodiments.

[0069] The following describes the computer-readable storage medium provided in the embodiments of this application. The computer-readable storage medium described below can be referred to in correspondence with the artificial intelligence-based human experience management method for traditional Chinese medicine described above.

[0070] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described artificial intelligence-based method for managing human experience in traditional Chinese medicine.

[0071] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0072] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0073] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A human experience management system for traditional Chinese medicine based on artificial intelligence, characterized in that, It includes a blockchain network consisting of public and permissioned blockchains and multiple edge computing nodes deployed in various medical institutions; Each of the aforementioned edge computing nodes is used to collect users' traditional Chinese medicine clinical data, which includes user profiles, traditional Chinese medicine prescription data, and efficacy evaluation data. The user's traditional Chinese medicine clinical data is encrypted using homomorphic encryption technology to obtain encrypted traditional Chinese medicine clinical data. Feature extraction is then performed on the encrypted traditional Chinese medicine clinical data using artificial intelligence technology to obtain feature information. The feature information is input into a pre-built traditional Chinese medicine analysis model to obtain the user's traditional Chinese medicine knowledge rule information, which includes the applicable disease range, optimal dosage range, contraindication information, and efficacy influencing factors of traditional Chinese medicine. The public blockchain is used to store the traditional Chinese medicine knowledge rule information sent by each edge computing node, and to determine multiple user experience information based on each of the traditional Chinese medicine knowledge rule information, and to construct a traditional Chinese medicine knowledge graph based on each of the user experience information. The permissioned chain is used to manage permissions for account information and nodes in the blockchain network. The account information includes doctors and regulators, and the permission information includes data access permissions, data modification permissions, and model training permissions.

2. The artificial intelligence-based traditional Chinese medicine human experience management system according to claim 1, characterized in that, Each of the edge computing nodes includes a medication dispensing module, which is specifically used for: Acquire user symptom information, symptom images, and user profiles, wherein the symptom information is either voice information or text information; The user's symptom information is segmented into words to obtain multiple keywords related to the symptom information; All the keywords are classified based on a preset text classification model to obtain the dimension to which each keyword belongs, and the dimension represents the domain of the corresponding keyword; Based on a preset semantic analysis technique, semantic analysis is performed on all the keywords to obtain a semantic description of each keyword, and the semantic description represents the medical terminology of the keyword; Based on a pre-set sensitive word library, the sensitivity level of each keyword is determined. The pre-set sensitive word library is set based on medical contraindications, social media public opinion hotspots, and public health emergency events. The symptom images are used to identify features based on image recognition technology to obtain symptom features; The semantic description and symptom features of each keyword are fused together, and a multimodal fusion method based on attention mechanism is used to obtain the symptom description text. Based on the user's corresponding symptom description text, the sensitivity of multiple keywords, and a pre-constructed traditional Chinese medicine knowledge graph, multiple initial prescription plans for the user are determined; Based on the user's historical medication information, user profile, and preset medication evaluation rules, each initial medication plan is evaluated to obtain an evaluation score for each initial medication plan; Each of the initial medication dispensing protocols and the evaluation score for each initial medication dispensing protocol are sent to the doctor's terminal so that the doctor can dispense medication for the user based on the evaluation scores.

3. The artificial intelligence-based traditional Chinese medicine human experience management system according to claim 1, characterized in that, The medication dispensing module, used to evaluate each initial medication dispensing plan based on the user's historical medication dispensing information, user profile, and preset medication dispensing evaluation rules, specifically uses the following to obtain an evaluation score for each initial medication dispensing plan: Obtain historical evaluation information for each of the initial drug preparation schemes; Based on the historical evaluation information of each initial medication regimen, a first weight value is determined for each initial medication regimen; Based on the user's symptom information, symptom images, user profile, and each of the initial medication regimens, the expected efficacy of each initial medication regimen is evaluated in the artificial intelligence model. Based on the expected efficacy of each of the initial medication regimens, a second weight value is determined for each of the initial medication regimens; Based on the degree of synergistic and antagonistic influence between each Chinese herb in each initial dispensing scheme, a third weight value is determined for each initial dispensing scheme; Based on the user's medical history, age, gender, allergy history, each initial medication regimen, and a preset neural network model in the user profile, the probability of adverse reactions occurring when the user uses each initial medication regimen is predicted; Based on the probability of adverse reactions for each initial medication regimen, a fourth weight value is determined for each initial medication regimen; A fifth weight value is determined for each initial medication regimen based on its price, efficacy, and the user's preferences in the user profile. Obtain the inventory information of each Chinese medicine corresponding to each initial dispensing plan from each medical institution, and determine the sixth weight value of each initial dispensing plan based on the inventory information of each Chinese medicine corresponding to each initial dispensing plan. An evaluation score is determined for each initial medication preparation scheme based on the first, second, third, fourth, fifth, and sixth weight values.

4. The artificial intelligence-based traditional Chinese medicine human experience management system according to claim 2, characterized in that, Each edge computing node also includes a traditional Chinese medicine clinical data acquisition module. Specifically, the traditional Chinese medicine clinical data acquisition module is used to: after the doctor prescribes medicine for the user, obtain the corresponding initial medication template based on the final prescription information. The medication template includes medical institution information, patient basic information, disease diagnosis information, prescription information, medication guidance information, and multiple time content nodes. The prescription information includes the name, dosage form, specifications, and quantity of traditional Chinese medicine. The time content nodes are used to record medication feedback information later. Based on the user profile, the initial medication template is adjusted to obtain a new medication template. The user profile includes lifestyle habits, past medical history, and allergy history. The new medication template is sent to the user's terminal so that the user can fill in medication feedback information in multiple time content nodes corresponding to the new medication template according to the actual medication situation, and generate a target medication template. The medication feedback information includes medication method and dosage, medication time, symptom pictures, description or consultation voice or text information, and one or more other medication information. After the user fills in the medication feedback information, the validity of the target medication template is verified. The validity verification includes data integrity verification, data rationality verification, and data consistency verification. If the validity of the user's target medication template is verified, the target medication template is preprocessed to obtain the user's traditional Chinese medicine clinical data.

5. The artificial intelligence-based traditional Chinese medicine human experience management system according to claim 1, characterized in that, Each of the aforementioned edge computing nodes, when constructing a traditional Chinese medicine analysis model, is specifically used for: Obtain a training set, which includes multiple homomorphically encrypted historical clinical experience information of traditional Chinese medicine. The historical clinical experience information of traditional Chinese medicine consists of changes in symptoms and signs of patients after medication, as well as time-series characteristics of drug efficacy, arranged in a time sequence. Based on a bidirectional LSTM network, the patient's symptom changes and signs after medication are arranged in time series as input, and the drug efficacy time series features are output. Based on graph convolutional neural networks, the traditional Chinese medicine knowledge graph is used as input and prescription compatibility features are used as output; Based on the outputs of the time-series pharmacodynamics channel and the compatibility analysis channel, the feature fusion weights of the two channels are dynamically adjusted through a gating mechanism, and the fused features are output. The patient's syndrome type, current season, and regional information are encoded to generate a dynamic adjustment coefficient matrix, which is used to dynamically adjust the weight parameters in the dual-channel neural network model. Initialize the dual-channel neural network model according to the training set, calculate the encryption gradient, and send the encryption gradient to the public chain; The public blockchain is used to obtain the encryption gradient of each edge node, perform a weighted average based on the dynamic aggregation weight to obtain the aggregated encryption gradient, and send the aggregated encryption gradient to each edge computing node. Each of the aforementioned edge computing nodes is also used to update the dual-channel neural network model based on the aggregated encrypted gradient, thereby constructing a traditional Chinese medicine analysis model.

6. The artificial intelligence-based traditional Chinese medicine human experience management system according to claim 1, characterized in that, The public blockchain, used to determine multiple users' experience information based on the various traditional Chinese medicine knowledge rules, is specifically used for: According to the categories of Chinese medicine and the types of diseases, the information on the rules of Chinese medicine knowledge is classified to obtain multiple Chinese medicine knowledge datasets; For any of the traditional Chinese medicine knowledge datasets, determine whether there are different traditional Chinese medicine knowledge rule information obtained by the same user on multiple edge computing nodes in the traditional Chinese medicine knowledge datasets; For any of the traditional Chinese medicine knowledge datasets, if there are different traditional Chinese medicine knowledge rule information obtained by the same user on multiple edge computing nodes, then the traditional Chinese medicine knowledge rules of the same patient on different edge computing nodes are associated and integrated to obtain new traditional Chinese medicine knowledge rule information. For any given set of traditional Chinese medicine knowledge datasets, multiple sets of human experience information on traditional Chinese medicine are obtained based on each traditional Chinese medicine knowledge rule in the dataset.

7. The artificial intelligence-based traditional Chinese medicine human experience management system according to claim 6, characterized in that, The public blockchain is also used for: When more than two pieces of human experience information are detected for the same traditional Chinese medicine knowledge point, the smart contract is triggered, and the human experience information for the traditional Chinese medicine knowledge point is updated according to the voting results of the smart contract. The smart contract is used to organize relevant medical institution nodes to vote on the same traditional Chinese medicine knowledge point using experience information from multiple people, obtain voting information, and determine the voting result based on each voting information and a preset voting weight calculation rule.

8. A method for managing human experience in using traditional Chinese medicine based on artificial intelligence, characterized in that, include: Collect users' clinical data on traditional Chinese medicine, including user profiles, traditional Chinese medicine prescription data, and efficacy evaluation data; The user's traditional Chinese medicine clinical data is encrypted using homomorphic encryption technology to obtain encrypted traditional Chinese medicine clinical data. Feature information is obtained by extracting features from the encrypted clinical data of traditional Chinese medicine based on artificial intelligence technology. The feature information is input into a pre-built traditional Chinese medicine analysis model to obtain the user's traditional Chinese medicine knowledge rule information, which includes the applicable disease range, optimal dosage range, contraindication information, and efficacy influencing factors of traditional Chinese medicine. The traditional Chinese medicine knowledge rules information is sent to a blockchain network for storage, so that the blockchain network can determine multiple user experience information based on the traditional Chinese medicine knowledge rules information sent by each edge computing node, and construct a traditional Chinese medicine knowledge graph based on each user experience information.

9. An electronic device, characterized in that, Includes a processor, which is coupled to a memory; The processor is configured to execute a computer program stored in the memory, so that the electronic device performs the method as described in claim 1.

10. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in claim 1.

Images