A method for automatically checking compatibility of traditional Chinese medicine prescriptions

Abstract

The application relates to the field of traditional Chinese medicine information and medical information processing technology, and discloses an automatic examination method for traditional Chinese medicine prescription compatibility contraindication, which comprises the following steps: S1, traditional Chinese medicine contraindication data acquisition and standardization stage; S2, compatibility contraindication knowledge graph construction stage; S3, rule engine knowledge base generation stage; S4, electronic prescription structured analysis stage; S5, prescription dose quantization preprocessing stage; S6, patient individualization information fusion stage; S7, multi-dimensional contraindication matching examination stage; S8, examination result output and recording stage; the application realizes automatic examination of traditional Chinese medicine prescription compatibility contraindication, replaces part of traditional manual examination links, solves the problems of non-uniform manual examination standards, high omission rate and low efficiency, can accurately identify the contraindication differences of traditional Chinese medicine aliases, different names and different processed products, and avoids misjudgment caused by simple keyword matching.

Description

Technical Field

[0001] This invention relates to the field of information technology in traditional Chinese medicine and medical information processing, and in particular to an automatic method for reviewing the incompatibilities of traditional Chinese medicine prescriptions. Background Technology

[0002] The compatibility of Chinese herbal prescriptions is a core aspect of clinical treatment in traditional Chinese medicine. Reasonable compatibility can enhance efficacy and reduce toxicity, while improper compatibility may cause adverse reactions or even endanger the patient's life. Traditional Chinese medicine prescription review mainly relies on pharmacists' personal experience. However, due to limitations in pharmacists' professional level, workload, and memory capacity, it is difficult to guarantee the consistency and comprehensiveness of the review results. With the advancement of hospital informatization, some medical institutions have begun to adopt electronic prescription review systems. However, existing systems are mostly based on simple keyword matching technology, which cannot adapt to the complex knowledge system and clinical application scenarios of traditional Chinese medicine, and cannot meet the dual needs of modern hospitals for prescription safety and efficiency. Summary of the Invention

[0003] The purpose of this invention is to provide an automatic method for reviewing the incompatibilities of traditional Chinese medicine prescriptions, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: An automatic method for reviewing incompatibilities in traditional Chinese medicine prescriptions includes the following steps: S1. Data collection and standardization of Chinese medicine contraindications: Collect original data on Chinese medicine contraindications from multiple authoritative sources, perform data cleaning and deduplication, and establish a two-way mapping system for standard Chinese medicine names. S2, the knowledge graph construction stage of drug compatibility and incompatibilities: define core entities and relationship types, add multi-dimensional attribute information to each relationship, and construct a structured knowledge graph of drug compatibility and incompatibilities. S3, the rule engine knowledge base generation stage, converts knowledge graph triples into executable rule expressions, establishes a rule priority system, and realizes the visual editing and dynamic updating of rules; S4. Electronic prescription structured parsing stage: Receive electronic prescription data transmitted from the hospital information system, extract core elements of the prescription, and complete the standardized conversion of Chinese medicine names; S5. Prescription dosage quantification preprocessing stage: Calculate the daily dosage of each Chinese herb, compare the calculation results with the dosage range specified in the pharmacopoeia, and generate dosage anomaly markers; S6. In the patient-personalized information fusion stage, extract TCM syndrome diagnosis information from electronic medical records, integrate the patient's special physiological and pathological information, and generate a patient-personalized contraindication feature vector. S7. In the multi-dimensional taboo matching review stage, the rule engine is called to execute multiple types of taboo matching reviews in sequence, and priority conflicts when multiple rules are triggered are detected and resolved. S8. In the stage of review result output and recording, the contraindications found are classified, detailed review instructions are generated, and the results are fed back to the doctor's workstation in real time and the processing process is recorded.

[0005] As a further improvement to this technical solution: Step S1 specifically includes: collecting original data on contraindications of traditional Chinese medicine from the Pharmacopoeia of the People's Republic of China, the Dictionary of Traditional Chinese Medicine, and authoritative clinical guidelines, and removing duplicate and invalid data; establishing a two-way mapping mechanism between the standard name of the pharmacopoeia and its alternative names, synonyms, and processed product names; maintaining a three-layer mapping table, with the first layer being a general alternative name mapping, the second layer being a regional alternative name mapping, and the third layer being a processing product-specific mapping.

[0006] As a further improvement to this technical solution: Step S2 specifically includes: constructing three core entities: traditional Chinese medicine entity, syndrome entity, and patient entity; defining five core relationships: antagonistic relationship, incompatible relationship, pregnancy contraindication relationship, syndrome contraindication relationship, and dosage contraindication relationship; and adding dosage threshold, applicable population, contraindication degree, and evidence level attribute information to each relationship.

[0007] As a further improvement to this technical solution, step S3 specifically includes: converting entity relation triples in the knowledge graph into machine-executable rule expressions; establishing a rule priority system based on taboo severity and evidence level; and developing a rule visualization editing interface that supports adding, modifying, and deleting rules without restarting the system.

[0008] As a further improvement to this technical solution, step S4 specifically includes: receiving electronic prescription data in XML or JSON format transmitted from the hospital information system through a standardized interface; extracting patient basic information, diagnostic information, list of Chinese medicines, single dose, total number of doses, and usage; calling the bidirectional mapping system of standard Chinese medicine names to complete the standardized conversion of prescription Chinese medicine names, and identifying and labeling the information of processed products.

[0009] As a further improvement to this technical solution: step S5 specifically includes: calculating the daily dosage of each Chinese herbal medicine, using the following formula:

[0010] in, This refers to the daily dosage of traditional Chinese medicine, in grams. This refers to the dosage of the traditional Chinese medicine in a single-dose prescription, in grams. This refers to the total number of doses in the prescription. The total number of days of administration specified in the prescription, in days; the calculated daily dose is compared with the commonly used dose range specified in the Pharmacopoeia of the People's Republic of China; when the daily dose exceeds the commonly used dose range, a dose abnormality marker is generated.

[0011] As a further improvement to this technical solution: Step S7 specifically includes: calling the rule engine to sequentially review the Eighteen Incompatibilities, Nineteen Antagonisms, Pregnancy Contraindications, Syndrome Contraindications, and Dosage Contraindications; detecting conflicts when multiple rules are triggered simultaneously, and determining the final review result according to the rule priority score. The rule priority score calculation formula is:

[0012] The rules are ranked in descending order of priority scores, and the rule with the highest score is selected as the final effective rule.

[0013] As a further improvement to this technical solution: Step S8 specifically includes: classifying the discovered contraindications into three levels according to their severity: prohibited, used with caution, and caution; generating a detailed description of each contraindication, including the drug involved, the type of contraindication, and the basis for the contraindication; feeding back the review results to the doctor's workstation using different colors; and recording the doctor's handling of the warning information, including whether to adopt and modify, ignore the warning, and the reasons.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention realizes the automated review of the incompatibilities of traditional Chinese medicine prescriptions, replacing some of the traditional manual review process. It solves the problems of inconsistent standards, high omission rate, and low efficiency in manual review. It can accurately identify the differences in contraindications of different names, synonyms, and processed products of traditional Chinese medicine, avoiding misjudgments caused by simple keyword matching. At the same time, by quantitatively calculating the daily dosage of traditional Chinese medicine, it achieves accurate judgment of dose-dependent contraindications and distinguishes the safety differences of drug combinations at different dosages.

[0015] 2. This invention separates contraindication rules from system code, supports visual editing and dynamic updates of rules, and allows for the addition, modification, and deletion of contraindication knowledge without restarting the system, reducing system maintenance costs and shortening the knowledge update cycle. It also enables personalized review by combining patients' TCM syndrome diagnosis and special physiological and pathological information, reducing over-warning and under-warning situations. Furthermore, by recording doctors' processing results of warning information, it can accumulate clinical application data, providing support for subsequent rule optimization and system iteration.

[0016] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it according to the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Specific embodiments of the present invention are given in detail below with reference to the accompanying drawings. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the structure of an automatic method for reviewing the incompatibilities of traditional Chinese medicine prescriptions. Detailed Implementation

[0018] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are for illustrative purposes only and are not intended to limit the scope of the invention. The invention is described more specifically in the following paragraphs by way of example with reference to the accompanying drawings. It should be noted that the drawings are in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0019] Please see Figure 1 In this embodiment of the invention, an automatic method for reviewing the incompatibilities of traditional Chinese medicine prescriptions includes the following steps: S1. Data collection and standardization of Chinese medicine contraindications: Collect original data on Chinese medicine contraindications from multiple authoritative sources, perform data cleaning and deduplication, and establish a two-way mapping system for standard Chinese medicine names. S2, the knowledge graph construction stage of drug compatibility and incompatibilities: define core entities and relationship types, add multi-dimensional attribute information to each relationship, and construct a structured knowledge graph of drug compatibility and incompatibilities. S3, the rule engine knowledge base generation stage, converts knowledge graph triples into executable rule expressions, establishes a rule priority system, and realizes the visual editing and dynamic updating of rules; S4. Electronic prescription structured parsing stage: Receive electronic prescription data transmitted from the hospital information system, extract core elements of the prescription, and complete the standardized conversion of Chinese medicine names; S5. Prescription dosage quantification preprocessing stage: Calculate the daily dosage of each Chinese herb, compare the calculation results with the dosage range specified in the pharmacopoeia, and generate dosage anomaly markers; S6. In the patient-personalized information fusion stage, extract TCM syndrome diagnosis information from electronic medical records, integrate the patient's special physiological and pathological information, and generate a patient-personalized contraindication feature vector. S7. In the multi-dimensional taboo matching review stage, the rule engine is called to execute multiple types of taboo matching reviews in sequence, and priority conflicts when multiple rules are triggered are detected and resolved. S8. In the stage of review result output and recording, the contraindications found are classified, detailed review instructions are generated, and the results are fed back to the doctor's workstation in real time and the processing process is recorded. Specifically, in the stage of collecting and standardizing data on contraindications of traditional Chinese medicine, the acquisition and standardization of basic data are completed, providing a standardized data source for the subsequent construction of knowledge graphs; In the stage of constructing the knowledge graph of compatibility taboos, the scattered taboo knowledge is transformed into a structured graph data structure to realize the visualization and reasoning of taboo relationships; In the knowledge base generation stage of the rule engine, the structured knowledge graph is converted into machine-executable rules, the priority logic for rule execution is established, and the dynamic maintenance of rules is supported. In the structured parsing stage of electronic prescriptions, data integration between the hospital information system and the review system is achieved, valid information in the prescriptions is extracted, and names are standardized. In the preprocessing stage of prescription dosage quantification, the standardized calculation and anomaly detection of traditional Chinese medicine dosage are completed, providing a quantitative basis for the review of dose-dependent contraindications; In the patient-personalized information integration stage, individual differences among patients are integrated to achieve personalized contraindication review based on the specific circumstances of each patient. In the multi-dimensional taboo matching review phase, the core taboo matching logic is executed to resolve conflicts when multiple rules are triggered simultaneously. In the review results output and recording stage, the review results are displayed and feedback is provided in a tiered manner, and the processing procedures are recorded for subsequent system optimization.

[0020] Step S1 specifically includes: collecting original data on contraindications of traditional Chinese medicine from the Pharmacopoeia of the People's Republic of China, the Dictionary of Traditional Chinese Medicine, and authoritative clinical guidelines, and removing duplicate and invalid data; establishing a two-way mapping mechanism between the standard names of pharmacopoeias and their alternative names, synonyms, and processed product names; and maintaining a three-layer mapping table, with the first layer being the mapping of general alternative names, the second layer being the mapping of regional alternative names, and the third layer being the mapping of processed product-specific names. Specifically, data is collected from the Pharmacopoeia of the People's Republic of China, the Dictionary of Traditional Chinese Medicine, and authoritative clinical guidelines to ensure the authority and accuracy of basic contraindication knowledge; Remove duplicate and invalid data to improve the purity of the knowledge base and avoid system misjudgments caused by redundant rules; Establish a two-way mapping mechanism between pharmacopoeia standard names and alternative names, synonyms, and processed product names. The forward mapping converts non-standard names in prescriptions into standard names that the system can recognize, while the reverse mapping supports the conversion of system output results into commonly used clinical names, ensuring the convenience of clinical use. The system maintains a three-layer mapping table. The first layer, the general alias mapping, covers the common aliases of Chinese medicines nationwide. The second layer, the regional alias mapping, adapts to the differences in medication habits in different regions. The third layer, the processing product-specific mapping, distinguishes the contraindications of different processing products of the same Chinese medicine, thus solving the problem of different contraindications for different processing products.

[0021] The S2 step specifically includes: constructing three core entities: Chinese medicine entity, syndrome entity, and patient entity; defining five core relationships: antagonistic relationship, incompatible relationship, pregnancy contraindication relationship, syndrome contraindication relationship, and dosage contraindication relationship; and adding attribute information such as dosage threshold, applicable population, degree of contraindication, and level of evidence to each relationship. Specifically, three core entities are constructed: traditional Chinese medicine entity, syndrome entity, and patient entity, covering all core objects involved in the incompatibilities of traditional Chinese medicine prescriptions; It defines five core relationships: incompatible relationships, antagonistic relationships, pregnancy contraindications, syndrome contraindications, and dosage contraindications, comprehensively covering common types of drug incompatibilities in clinical practice of traditional Chinese medicine; Each relationship is assigned attributes including dosage threshold, applicable population, contraindication level, and evidence level. The dosage threshold is used to distinguish contraindication at different dosages, the applicable population is used to limit the scope of application of the contraindication relationship, the contraindication level is used to classify the warning level, and the evidence level is used to assess the reliability of the contraindication knowledge, providing a basis for setting rule priorities.

[0022] The S3 steps specifically include: converting entity relation triples in the knowledge graph into machine-executable rule expressions; establishing a rule priority system based on taboo severity and evidence level; and developing a visual rule editing interface that supports adding, modifying, and deleting rules without restarting the system. Specifically, the entity relation triples in the knowledge graph are converted into machine-executable rule expressions, realizing the connection between the knowledge graph and the rule engine, so that structured knowledge can be automatically executed by the computer; Establish a rule priority system based on the severity of taboos and the level of evidence to ensure that high-risk, high-credibility taboo rules are executed first, and to avoid low-priority rules interfering with the detection of core risks; Develop a visual rule editing interface that supports adding, modifying, and deleting rules without restarting the system, reducing system maintenance costs and enabling rapid updates and iterations of taboo knowledge.

[0023] The S4 step specifically includes: receiving electronic prescription data in XML or JSON format transmitted from the hospital information system through a standardized interface; extracting basic patient information, diagnostic information, list of Chinese medicines, single dose dosage, total number of doses, and usage instructions; calling the bidirectional mapping system for standard Chinese medicine names to complete the standardized conversion of prescription Chinese medicine names, and identifying and labeling information on processed products; Specifically, it receives electronic prescription data in XML or JSON format from hospital information systems through standardized interfaces to ensure compatibility between different hospital information systems and the review system; Extract patient basic information, diagnostic information, list of Chinese medicines, single dose dosage, total number of doses and usage instructions to obtain all the core information required for prescription review; By invoking the bidirectional mapping system of standard Chinese medicine names, the standardized conversion of prescription Chinese medicine names is completed, and the information of processed products is identified and labeled to ensure that subsequent contraindication reviews can accurately distinguish between different Chinese medicines and different processed products.

[0024] Step S5 specifically includes: calculating the daily dosage of each Chinese herbal medicine, using the following formula:

[0025] in, This refers to the daily dosage of traditional Chinese medicine, in grams. This refers to the dosage of the traditional Chinese medicine in a single-dose prescription, in grams. This refers to the total number of doses in the prescription. The total number of days of administration specified in the prescription, in days; compare the calculated daily dose with the commonly used dose range specified in the Pharmacopoeia of the People's Republic of China; when the daily dose exceeds the commonly used dose range, a dose abnormality marker is generated; Specifically, it clarifies the calculation method, comparison standards, and rules for marking abnormalities for daily doses; The daily dosage of each Chinese herbal medicine is calculated using the following formula: The formula is used to convert the single-dose dosage in a prescription into the actual daily dosage taken by the patient, thus solving the problem of incomparable dosages caused by differences in the total number of doses and the number of days of administration in different prescriptions. in, This refers to the daily dosage of traditional Chinese medicine, in grams. This refers to the dosage of the traditional Chinese medicine in a single-dose prescription, in grams. This refers to the total number of doses in the prescription. The total number of days to be taken as prescribed, in days; The calculated daily dose is compared with the commonly used dose range specified in the Pharmacopoeia of the People's Republic of China, and the national legal standard is used as the basis for judging abnormal doses. When the daily dose exceeds the commonly used dose range, a dose abnormality marker is generated, providing a trigger condition for subsequent dose contraindication review.

[0026] Step S7 specifically includes: calling the rule engine to sequentially review the Eighteen Incompatibilities, Nineteen Antagonisms, Pregnancy Contraindications, Syndrome Contraindications, and Dosage Contraindications; detecting conflicts when multiple rules are triggered simultaneously; and determining the final review result based on rule priority scoring. The rule priority scoring formula is as follows:

[0027] The rules are ranked in descending order of priority scores, and the rule with the highest score is selected as the final effective rule. Specifically, it clarifies the order of taboo review, conflict resolution methods, and priority scoring calculation methods; The rule engine is invoked to sequentially review the Eighteen Incompatibilities, Nineteen Antagonisms, Contraindications during Pregnancy, Contraindications due to Syndrome, and Contraindications due to Dosage, in order of severity from highest to lowest, to ensure that high-risk contraindications are detected first. The system detects conflicts when multiple rules are triggered simultaneously, and determines the final review result based on rule priority scores. The rule priority score calculation formula is as follows: The formula is used to quantitatively evaluate the priority of each rule, taking into account both the severity of the taboo and the reliability of the evidence, and to solve the problem of execution order when multiple rules are triggered simultaneously. in, Score the rule priority. Assign a value to the severity of the taboo. Assign a value to the level of evidence. This is a severity weighting coefficient. This refers to the weighting coefficient for the level of evidence. The rules are ranked in descending order of priority scores, and the rule with the highest score is selected as the final effective rule to ensure that the most critical contraindication information is presented to doctors first.

[0028] The S8 steps specifically include: classifying the discovered contraindications into three levels according to their severity: prohibited, use with caution, and caution; generating a detailed description of each contraindication, including the drug involved, the type of contraindication, and the basis for the contraindication; feeding back the review results to the doctor's workstation using different colors; and recording the doctor's handling of the warning information, including whether to adopt and modify the warning, ignore the warning, and the reasons. Specifically, contraindications are classified into three levels according to their severity: prohibited, used with caution, and caution, enabling doctors to quickly identify issues of different risk levels. Generate detailed descriptions for each contraindication question, including the drug involved, the type of contraindication, and the basis for the contraindication, providing doctors with sufficient reference information to determine whether to modify the prescription; The review results are fed back to the doctor's workstation using different colors to visually distinguish and improve the doctor's attention and the efficiency of the communication of early warning information. Record doctors' handling of warning information, including whether to adopt or modify the warning, ignore the warning, and provide reasons, accumulate clinical application data, and provide a basis for subsequent rule optimization and system iteration.

[0029] The method of use and working principle of this invention are as follows: Usage: First, collect and standardize the knowledge of contraindications of traditional Chinese medicine (TCM), construct a knowledge graph of TCM compatibility contraindications containing multi-dimensional attributes, and convert it into a dynamically updatable rule engine knowledge base. When the system is running, it receives electronic prescription data transmitted from the hospital information system, extracts the core elements of the prescription, completes the standardized conversion of TCM names, calculates the daily dosage of TCM and compares it with the pharmacopoeia standard, and integrates the patient's TCM syndrome diagnosis and physiological and pathological information to generate personalized contraindication feature vectors. It calls the rule engine to sequentially perform multi-type contraindication matching reviews, detects and resolves priority conflicts when multiple rules are triggered simultaneously, and finally classifies the discovered contraindication issues, generates detailed review descriptions and feeds them back to the doctor's workstation, while recording the doctor's processing of the warning information.

[0030] Working principle: The system uses knowledge graph technology to structure and represent the scattered knowledge of contraindications for traditional Chinese medicine (TCM) combinations, establishing relationships between TCM, syndromes, and patients, and adding multi-dimensional attribute information to each contraindication relationship. Then, a rule engine converts the structured knowledge graph into machine-executable rule expressions, establishing a rule priority system based on the severity of contraindications and the level of evidence. When an electronic prescription is received, the system first parses and preprocesses the prescription information, converting non-standard prescription data into a standardized format that the system can recognize. Then, combined with the patient's personalized information, the rule engine performs multi-dimensional contraindication matching on the prescription. When multiple rules are triggered simultaneously, a priority scoring mechanism determines the final effective rule, thereby achieving automated and intelligent review of TCM prescription contraindications.

[0031] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Those skilled in the art can readily implement the present invention based on the description and drawings above. However, any modifications, alterations, and variations made by those skilled in the art without departing from the scope of the present invention using the disclosed technical content are equivalent embodiments of the present invention. Furthermore, any modifications, alterations, and variations made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the present invention.

Claims

1. A method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions, characterized in that, Includes the following steps: S1. Data collection and standardization of Chinese medicine contraindications: Collect original data on Chinese medicine contraindications from multiple authoritative sources, perform data cleaning and deduplication, and establish a two-way mapping system for standard Chinese medicine names. S2, the knowledge graph construction stage of drug compatibility and incompatibilities: define core entities and relationship types, add multi-dimensional attribute information to each relationship, and construct a structured knowledge graph of drug compatibility and incompatibilities. S3, the rule engine knowledge base generation stage, converts knowledge graph triples into executable rule expressions, establishes a rule priority system, and realizes the visual editing and dynamic updating of rules; S4. Electronic prescription structured parsing stage: Receive electronic prescription data transmitted from the hospital information system, extract core elements of the prescription, and complete the standardized conversion of Chinese medicine names; S5. Prescription dosage quantification preprocessing stage: Calculate the daily dosage of each Chinese herb, compare the calculation results with the dosage range specified in the pharmacopoeia, and generate dosage anomaly markers; S6. In the patient-personalized information fusion stage, extract TCM syndrome diagnosis information from electronic medical records, integrate the patient's special physiological and pathological information, and generate a patient-personalized contraindication feature vector. S7. In the multi-dimensional taboo matching review stage, the rule engine is called to execute multiple types of taboo matching reviews in sequence, and priority conflicts when multiple rules are triggered are detected and resolved. S8. In the stage of review result output and recording, the contraindications found are classified, detailed review instructions are generated, and the results are fed back to the doctor's workstation in real time and the processing process is recorded.

2. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The S1 step specifically includes: collecting original data on contraindications of traditional Chinese medicine from the Pharmacopoeia of the People's Republic of China, the Dictionary of Traditional Chinese Medicine, and authoritative clinical guidelines, and removing duplicate and invalid data; establishing a two-way mapping mechanism between the standard names in the pharmacopoeia and their alternative names, synonyms, and processed product names; and maintaining a three-layer mapping table, with the first layer being a general alternative name mapping, the second layer being a regional alternative name mapping, and the third layer being a processing product-specific mapping.

3. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The S2 step specifically includes: constructing three core entities: traditional Chinese medicine entity, syndrome entity, and patient entity; defining five core relationships: antagonistic relationship, mutual restraint relationship, pregnancy contraindication relationship, syndrome contraindication relationship, and dosage contraindication relationship; and adding attribute information such as dosage threshold, applicable population, degree of contraindication, and level of evidence to each relationship.

4. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The S3 step specifically includes: converting entity relation triples in the knowledge graph into machine-executable rule expressions; establishing a rule priority system based on taboo severity and evidence level; and developing a rule visualization editing interface that supports adding, modifying, and deleting rules without restarting the system.

5. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The S4 step specifically includes: receiving electronic prescription data in XML or JSON format transmitted from the hospital information system through a standardized interface; extracting patient basic information, diagnostic information, list of Chinese medicines, single dose, total number of doses and usage instructions; calling the bidirectional mapping system of standard Chinese medicine names to complete the standardized conversion of prescription Chinese medicine names, and identifying and labeling the information of processed products.

6. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, Step S5 specifically includes: calculating the daily dosage of each Chinese herbal medicine, using the following formula: in, This refers to the daily dosage of traditional Chinese medicine, in grams. This refers to the dosage of the traditional Chinese medicine in a single-dose prescription, in grams. This refers to the total number of doses in the prescription. The total number of days of administration specified in the prescription, in days; the calculated daily dose is compared with the commonly used dose range specified in the Pharmacopoeia of the People's Republic of China; when the daily dose exceeds the commonly used dose range, a dose abnormality marker is generated.

7. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The S7 step specifically includes: calling the rule engine to sequentially review the Eighteen Incompatibilities, Nineteen Antagonisms, Pregnancy Contraindications, Syndrome Contraindications, and Dosage Contraindications; detecting conflicts when multiple rules are triggered simultaneously; and determining the final review result based on rule priority scoring. The rule priority scoring formula is as follows: in, Score the rule priority. Assign a value to the severity of the taboo. Assign a value to the level of evidence. This is a severity weighting coefficient. This represents the weighting coefficient for the level of evidence; the rule priority scores are sorted in descending order, and the rule with the highest score is selected as the final effective rule.

8. The method for automatically reviewing the incompatibilities of traditional Chinese medicine prescriptions according to claim 1, characterized in that, The S8 step specifically includes: classifying the discovered contraindications into three levels according to their severity: prohibited, used with caution, and caution; generating a detailed description of each contraindication, including the drug involved, the type of contraindication, and the basis for the contraindication; feeding back the review results to the doctor's workstation using different colors; and recording the doctor's handling of the warning information, including whether to adopt and modify, ignore the warning, and the reasons.

Images