An intelligent decision-making diagnosis and treatment method, system and device combining traditional Chinese and Western medicine

Abstract

The application relates to the technical field of diagnosis and treatment, in particular to an intelligent decision-making diagnosis and treatment method, system and equipment combining traditional Chinese medicine and western medicine. First, the symptoms of a patient are encoded into a feature vector, then vector matching is performed in a pre-constructed knowledge graph combining traditional Chinese medicine and western medicine, applicable diagnosis and treatment information is screened out by calculating the similarity and setting a threshold value, and a preliminary scheme is formed; then the individual information of the patient is read, and analysis is performed in combination with a medication contraindication rule library, so that personalized treatment parameters are obtained; finally, the preliminary scheme and the personalized parameters are optimized and adjusted, and the final diagnosis and treatment suggestion is generated; not only the deep integration of traditional Chinese medicine and western medicine knowledge is realized, but also the diagnosis and treatment scheme can be dynamically adjusted according to the individual characteristics of the patient, and the accuracy and effectiveness of the diagnosis and treatment are improved.

Description

Technical Field

[0001] This application relates to the technical field of diagnosis and treatment, and in particular to an intelligent decision-making diagnosis and treatment method, system and equipment that combines traditional Chinese and Western medicine. Background Technology

[0002] Traditional Chinese medicine (TCM) and Western medicine each have their own unique characteristics and advantages. Integrating TCM and Western medicine in diagnosis and treatment can complement each other's strengths and improve treatment outcomes. With the development of medical informatization, applying artificial intelligence technology to the decision-making process in integrated TCM and Western medicine has become an important research direction. How to achieve the intelligent integration of TCM and Western medical knowledge and the formulation of personalized treatment plans is a crucial issue currently facing the medical field.

[0003] Currently, rule-based expert systems and case-based reasoning systems are used to assist in diagnosis and treatment. These systems recommend diagnostic and treatment plans through pre-set rule bases or historical case databases, which can assist doctors in making diagnostic and treatment decisions to a certain extent. At the same time, there have also been some attempts to apply knowledge graph technology in the medical field.

[0004] However, existing systems often treat traditional Chinese medicine and Western medicine knowledge separately, lacking an effective integration mechanism; and fail to fully consider individual patient differences, making it difficult to provide personalized treatment plans; this situation needs further improvement. Summary of the Invention

[0005] To address the problems that existing systems often treat traditional Chinese medicine and Western medicine knowledge separately, lacking an effective integration mechanism, and fail to fully consider individual patient differences, making it difficult to provide personalized treatment plans, this application provides an intelligent decision-making treatment method, system, and device that integrates traditional Chinese and Western medicine, employing the following technical solution:

[0006] Firstly, this application provides an intelligent decision-making and treatment method integrating traditional Chinese and Western medicine, comprising the following steps:

[0007] Receive patient symptom input, encode the symptoms, and calculate the symptom feature vector;

[0008] Based on the symptom feature vector, vector matching is performed in a preset integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph to calculate the similarity and filter out diagnosis and treatment information that exceeds a preset threshold to obtain a preliminary diagnosis and treatment plan.

[0009] Read individual patient information, perform feature analysis, and calculate personalized treatment parameters for the patient based on a pre-set medication contraindication rule base;

[0010] Based on the preliminary treatment plan and the patient's personalized treatment parameters, the plan is optimized to generate a final integrated traditional Chinese and Western medicine treatment recommendation.

[0011] By adopting the above technical solution, this application first encodes the patient's symptoms into feature vectors, then performs vector matching in a pre-constructed knowledge graph of integrated traditional Chinese and Western medicine, calculates similarity and sets thresholds to filter out applicable treatment information, forming a preliminary plan; next, it reads the patient's individual information and analyzes it in conjunction with a drug contraindication rule base to obtain personalized treatment parameters; finally, it optimizes and adjusts based on the preliminary plan and personalized parameters to generate the final treatment recommendations; this not only achieves a deep integration of traditional Chinese and Western medicine knowledge, but also dynamically adjusts the treatment plan according to the patient's individual characteristics, improving the accuracy and effectiveness of diagnosis and treatment.

[0012] Optionally, similarity is calculated and diagnostic information exceeding a preset threshold is filtered to obtain a preliminary treatment plan, specifically including the following steps:

[0013] The cosine similarity between the symptom feature vector and the diagnosis and treatment node vector in the knowledge graph is calculated to obtain a similarity score sequence.

[0014] Based on the similarity score sequence, threshold filtering is performed to select diagnostic and treatment nodes with similarity greater than a preset threshold, thereby obtaining a set of candidate diagnostic and treatment plans.

[0015] Based on the set of candidate treatment plans, the plans are fused, and each candidate plan is weighted and integrated according to the weight of the similarity score to obtain a preliminary treatment plan.

[0016] By adopting the above technical solution, this application first calculates the cosine similarity between the patient's symptom feature vector and all diagnosis and treatment node vectors stored in the knowledge graph, obtaining a similarity score sequence. These scores reflect the degree of matching between the current symptoms and historical diagnosis and treatment cases. Then, a similarity threshold is set, and diagnosis and treatment nodes with similarity higher than the threshold are selected to form a candidate solution set. Finally, these candidate solutions are fused, and weighted integration is performed based on the similarity score of each solution as a weight, resulting in a more comprehensive and accurate preliminary diagnosis and treatment plan. Through vectorization representation and similarity calculation, the limitations of traditional matching methods are effectively overcome, and the accuracy and flexibility of symptom matching are improved.

[0017] Optionally, the candidate treatment plan set includes a subset of traditional Chinese medicine treatment plans and a subset of Western medicine treatment plans. The plans are then fused, specifically including the following steps:

[0018] For the prescriptions in the subset of the TCM treatment plan, formula analysis is performed to extract the dosage and ratio information of the main and auxiliary drugs, and the compatibility characteristics of TCM are obtained.

[0019] Drug analysis is performed on the prescriptions in the subset of Western medicine treatment protocols to extract drug targets and dosage information, thereby obtaining the characteristics of Western medicine use.

[0020] Based on a pre-defined database of rules for combined use of traditional Chinese and Western medicines, compatibility checks are performed, and the interaction between the compatibility characteristics of traditional Chinese medicines and the characteristics of Western medicines is analyzed to obtain a coordinated medication plan.

[0021] Based on the aforementioned medication coordination plan, the plans are integrated, the ratio and dosage of Chinese and Western medicines are adjusted, and a preliminary treatment plan is generated.

[0022] By adopting the above technical solution, this application first analyzes the prescription of traditional Chinese medicine (TCM) to extract the dosage and ratio information of each herb (principal, assistant, adjuvant, and guide) to obtain complete TCM compatibility characteristics. Simultaneously, it analyzes Western medicine prescriptions to extract key information such as drug targets and dosages to obtain Western medicine usage characteristics. Then, based on a pre-established TCM-Western medicine combined use rule library, it checks the compatibility of these characteristics, analyzes potential interactions between TCM and Western medicines, and forms a coordinated medication plan. Finally, it optimizes and adjusts the ratio and dosage of TCM and Western medicines according to the coordinated plan to generate a preliminary treatment plan that maintains their respective characteristics while achieving synergistic effects. Through systematic analysis of TCM and Western medicine characteristics and the establishment of compatibility rules, the scientific compatibility of TCM and Western medicines is achieved, avoiding the problems that may arise from simple superposition.

[0023] Optionally, based on a pre-defined database of medication contraindications, personalized treatment parameters for the patient are calculated, including the following steps:

[0024] Based on a pre-defined database of contraindications for medication, standard medication parameters are extracted and used as baseline treatment parameters.

[0025] Read individual patient information to obtain tolerance and corresponding safety threshold parameters for different drug treatments;

[0026] Based on the patient's historical medication records, the therapeutic characteristics and dosage adjustment range of the corresponding drugs are determined. Combined with the tolerability of the different therapeutic effects of the drugs and the corresponding safety threshold parameters, the target dosage and dynamic drug control parameters are determined. The dynamic drug control parameters include dosage range parameters based on different therapeutic effects.

[0027] Based on the target dosage and the dynamic drug control parameters, personalized treatment parameters for the patient are obtained.

[0028] By adopting the above technical solution, this paper addresses the problem that existing systems often use uniform standards when setting medication parameters, failing to fully consider individual patient differences and medication history. For example, in treating hypertension, even with the same antihypertensive drug, different patients may have significantly different tolerance levels and response intensities. Simply applying a standard dosage may lead to poor treatment effects or adverse reactions. This application first extracts standard medication parameters from a medication contraindication rule library as benchmark parameters. Then, it reads the patient's individual information, including age, weight, liver and kidney function, to analyze the patient's tolerance to the therapeutic effects of different drugs and determine the corresponding safety threshold parameters. Next, by analyzing the patient's historical medication records, it extracts the actual therapeutic effects and appropriate dosage ranges of each drug. Combining this information with the patient's tolerance and safety thresholds, it calculates the most suitable target dosage and determines dynamic control parameters such as the dosage range for different therapeutic effects. Finally, based on these calculation results, a complete personalized treatment parameter system is formed. Through multi-dimensional analysis and dynamic adjustment mechanisms, the treatment plan not only conforms to medical standards but also adapts to individual characteristics, improving the accuracy and safety of medication.

[0029] Optionally, the dynamic drug control parameters include thresholds for limiting different drug interactions; after obtaining the patient's personalized treatment parameters based on the target dosage and the dynamic drug control parameters, the following steps are also included:

[0030] Based on the different therapeutic effects of drugs, several dose control units are generated within each administration range;

[0031] In the preset dynamic control drug delivery model, each dose control unit is used as a basic unit, and a target dosing value is set for all therapeutic drugs in the dose control unit. The sum of the effects of all therapeutic drugs in the dose control unit is set to be equal to the preset therapeutic effect threshold.

[0032] Based on the identified current drug combination and current dosing regimen, the corresponding drug interaction limitation threshold is obtained; taking each dose control unit as a basic unit, the interaction intensity of all therapeutic drugs in each dose control unit is controlled to be less than or equal to the corresponding limitation threshold.

[0033] By adopting the above technical solution, existing systems often only focus on the dosage at a single time point in drug administration control, neglecting the cumulative effect and mutual influence of drugs at different time periods. For example, in the treatment of chronic pain, if the dosage of analgesics and adjuvant drugs at different times of the day is not reasonably allocated, it may lead to excessive or insufficient efficacy at certain times, affecting the overall treatment effect. This application first divides multiple time periods as dose control units within their respective dosing ranges based on the therapeutic characteristics of different drugs, such as three dosing periods in the morning, noon, and evening. Then, in the dynamic control dosing model, the specific dosage of all therapeutic drugs in each time period is set, and it is ensured that the comprehensive effect intensity of all drugs in that time period can reach the expected therapeutic effect threshold. At the same time, based on the characteristics of the currently used drug combination, the interaction limit threshold between these drugs is obtained, and the interaction intensity of all drugs in each time period is strictly controlled to not exceed the safety limit. Through time-series precise dosing control and multiple safety constraints, the balanced distribution of drug effects and effective control of interactions are achieved, improving the safety of drug administration and the stability of treatment effects.

[0034] Optionally, based on the preliminary treatment plan and the patient's personalized treatment parameters, the plan is optimized to generate a final integrated traditional Chinese and Western medicine treatment recommendation, specifically including the following steps:

[0035] Based on the preliminary treatment plan, the plan was analyzed, and the information on traditional Chinese medicine prescriptions and Western medicine medications was extracted to obtain the components of the treatment plan.

[0036] Based on the patient's personalized treatment parameters, the dosage is adjusted, and the medication dosage in the components of the treatment plan is optimized to obtain the adjusted medication plan.

[0037] Based on the dosing range parameter and drug interaction limitation threshold in the medication dynamic control parameters, a constraint test is performed to verify whether the adjusted medication regimen meets the constraint conditions, and the verification results are obtained.

[0038] Based on the verification results, a treatment plan is generated, and the adjusted medication plan is integrated to generate the final integrated traditional Chinese and Western medicine treatment recommendations.

[0039] By adopting the above technical solution, this application first conducts a detailed analysis of the preliminary treatment plan, extracting the information on traditional Chinese medicine prescriptions and Western medicine dosages to form clear components of the treatment plan; then, based on the obtained personalized treatment parameters for the patient, it performs precise optimization calculations on the dosage of each component, including adjusting the dosage of each herb in the traditional Chinese medicine and the dosage of the Western medicine, to form a preliminary adjustment plan; next, according to the dosing range and drug interaction limit thresholds set in the dynamic control parameters for medication, it conducts a rigorous constraint test on the adjusted plan to ensure that all medication parameters are within a safe range, and records the verification results; finally, based on the verification results, it integrates the plans that pass the test to form a final integrated traditional Chinese and Western medicine treatment recommendation that considers individual characteristics and ensures safety and effectiveness.

[0040] Secondly, this application provides an intelligent decision-making and diagnosis system integrating traditional Chinese and Western medicine, comprising:

[0041] The symptom input module is used to receive patient symptom input, encode the symptoms, and calculate the symptom feature vector.

[0042] The treatment plan matching module is used to perform vector matching in a preset integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph based on the symptom feature vector, calculate the similarity and filter out diagnosis and treatment information that exceeds a preset threshold to obtain a preliminary diagnosis and treatment plan.

[0043] The parameter calculation module is used to read individual patient information, perform feature analysis, and calculate personalized treatment parameters for patients based on a preset medication contraindication rule library.

[0044] The treatment plan optimization module is used to optimize the treatment plan based on the preliminary treatment plan and the patient's personalized treatment parameters, and generate the final integrated traditional Chinese and Western medicine treatment recommendations.

[0045] Optionally, the scheme matching module includes:

[0046] The similarity calculation unit is used to perform cosine similarity calculation between the symptom feature vector and the diagnosis and treatment node vector in the knowledge graph to obtain a similarity score sequence.

[0047] The filtering unit is used to perform threshold filtering based on the similarity score sequence, filter out the diagnosis and treatment nodes with similarity greater than a preset threshold, and obtain a set of candidate diagnosis and treatment plans.

[0048] The scheme fusion unit is used to perform scheme fusion based on the set of candidate treatment schemes, and to integrate each candidate scheme by weighting the similarity scores to obtain a preliminary treatment scheme.

[0049] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned intelligent decision-making and diagnosis method combining traditional Chinese and Western medicine.

[0050] In summary, this application includes at least one of the following beneficial technical effects:

[0051] 1. This application first encodes patient symptoms into feature vectors, then performs vector matching in a pre-constructed knowledge graph of integrated traditional Chinese and Western medicine, calculates similarity and sets thresholds to filter applicable treatment information, forming a preliminary plan; next, it reads the patient's individual information and analyzes it in conjunction with a medication contraindication rule base to obtain personalized treatment parameters; finally, it optimizes and adjusts the preliminary plan and personalized parameters to generate the final treatment recommendations; this not only achieves a deep integration of traditional Chinese and Western medicine knowledge, but also dynamically adjusts the treatment plan according to the patient's individual characteristics, improving the accuracy and effectiveness of diagnosis and treatment;

[0052] 2. This application first analyzes traditional Chinese medicine (TCM) prescriptions, extracting the dosage and ratio information of each herb (principal, assistant, adjuvant, and guide) to obtain complete TCM compatibility characteristics. Simultaneously, it analyzes Western medicine prescriptions, extracting key information such as drug targets and dosages to obtain Western medicine usage characteristics. Then, based on a pre-established TCM-Western medicine combined use rule base, it checks the compatibility of these characteristics, analyzes potential interactions between TCM and Western medicines, and formulates a coordinated medication plan. Finally, based on the coordinated plan, it optimizes and adjusts the ratios and dosages of TCM and Western medicines to generate a preliminary treatment plan that maintains their respective characteristics while achieving synergistic effects. Through systematic analysis of TCM and Western medicine characteristics and the establishment of compatibility rules, it achieves scientific compatibility between TCM and Western medicines, avoiding the problems that may arise from simple superposition.

[0053] 3. This application first extracts standard medication parameters from a medication contraindication rule database as baseline parameters; then, it reads the patient's individual information, including age, weight, liver and kidney function, analyzes the patient's tolerance to different drug treatments, and determines the corresponding safety threshold parameters; next, by analyzing the patient's historical medication records, it extracts the actual therapeutic effect and appropriate dosage range of each drug, combines this information with the patient's tolerance and safety thresholds, calculates the most suitable target dosage, and determines dynamic control parameters such as the dosage range for different therapeutic effects; finally, based on these calculation results, a complete personalized treatment parameter system is formed; through multi-dimensional analysis and dynamic adjustment mechanisms, the treatment plan not only conforms to medical norms but also adapts to individual characteristics, improving the accuracy and safety of medication. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating an intelligent decision-making and treatment method combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0055] Figure 2 This is a flowchart illustrating step S200 in an intelligent decision-making and treatment method combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0056] Figure 3 This is a flowchart illustrating step S230 in an intelligent decision-making and treatment method combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0057] Figure 4 This is a flowchart illustrating step S300 in an intelligent decision-making and treatment method combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0058] Figure 5 This is a flowchart illustrating step S340 in an intelligent decision-making and treatment method combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0059] Figure 6 This is a flowchart illustrating step S400 in an intelligent decision-making and treatment method combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0060] Figure 7 This is a schematic diagram of a module of an intelligent decision-making and diagnosis system combining traditional Chinese and Western medicine, according to an embodiment of this application.

[0061] Figure 8 This is an internal structural diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0062] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0063] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0064] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0065] Firstly, this application provides an intelligent decision-making and treatment method that integrates traditional Chinese and Western medicine, referring to... Figure 1 It includes the following steps:

[0066] S100: Receive patient symptom input, encode the symptoms, and calculate the symptom feature vector.

[0067] In this embodiment, symptom coding refers to converting the symptom information described by the patient into a standardized digital code form; symptom descriptive words refer to the patient's colloquial expression of their own symptoms; standard terminology refers to standardized medical terminology; and symptom feature vector refers to a multi-dimensional numerical vector composed of symptom type, symptom severity, location of occurrence, and duration.

[0068] Specifically, a standardized symptom mapping database is established, containing three basic tables: a standard symptom terminology table, a symptom descriptive word mapping table, and an anatomical location coding table. When a patient inputs a symptom description, the system first converts the colloquial description into standard terminology using the symptom descriptive word mapping table, then assigns a severity value based on the symptom severity quantification standard, determines the location code using the anatomical location coding table, and finally assigns a time value based on the duration.

[0069] S200. Based on the symptom feature vector, perform vector matching in the preset integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph, calculate the similarity and filter out diagnosis and treatment information that exceeds the preset threshold to obtain a preliminary diagnosis and treatment plan.

[0070] In this embodiment, the integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph refers to a three-layer structured knowledge base that includes disease diagnosis, treatment plan, and medication advice; vector matching refers to using the cosine similarity calculation method to calculate the similarity between the input vector and the existing case vectors in the knowledge base; the diagnosis and treatment information includes three parts: Western medicine diagnosis, traditional Chinese medicine syndrome differentiation, and treatment plan advice.

[0071] Specifically, a knowledge graph of integrated traditional Chinese and Western medicine diagnosis and treatment for various common diseases is constructed. Each disease node includes standardized symptom feature vectors, diagnostic criteria, and treatment plans. The cosine similarity algorithm is used to calculate the similarity between the input symptom vector and the disease node vectors in the knowledge graph, selecting disease nodes with a similarity greater than 0.8 as candidate diagnoses. For each candidate diagnosis, its corresponding treatment plan is extracted as preliminary suggestions. For example, for headache symptoms, the system may match three diagnostic nodes: migraine, increased intracranial pressure, and cervical spondylosis, extracting corresponding Western medicine treatment plans and traditional Chinese medicine treatment suggestions for each.

[0072] S300: Read the patient's individual information, perform feature analysis, and calculate the patient's personalized treatment parameters based on the preset medication contraindication rule base.

[0073] In this embodiment, the patient's individual information includes four dimensions: basic information (age, gender, weight, height), medical history (past diseases, surgical records, family medical history), allergy information (drug allergies, food allergies), and medication records (recent medication use, long-term medication use). The medication contraindication rule base refers to a structured database containing drug use rules, which records the usage restrictions of various commonly used drugs. The personalized treatment parameters include specific values ​​for four aspects: dosage coefficient (adjustment coefficient of 0.5-1.5), medication interval (1-4 times per day), treatment course recommendation (3-30 days), and monitoring frequency (1-3 times per day).

[0074] Specifically, a rule-based medication parameter calculation system is established, comprising three core rule tables: an age-group medication rule table (dividing patients into three groups: under 14 years old, 15-64 years old, and over 65 years old), a special population rule table (pregnant women, breastfeeding women, and patients with hepatic or renal insufficiency), and a drug dosage adjustment table (setting a base dose based on a weight range of 40-90 kg). After reading patient information, the system first determines the patient's population category, then queries the corresponding medication rule, calculates the base dose based on weight parameters, and finally makes fine adjustments based on past medication response records.

[0075] S400 optimizes the initial treatment plan and the patient's personalized treatment parameters to generate the final integrated traditional Chinese and Western medicine treatment recommendations.

[0076] In this embodiment, the optimization of the treatment plan refers to the process of making personalized adjustments to the preliminary treatment plan; the adjustments include medication selection, dosage setting, medication time, and treatment cycle; the integrated traditional Chinese and Western medicine treatment suggestion refers to the final treatment plan.

[0077] Specifically, the system first conducts a safety review of the preliminary treatment plan to rule out any potential contraindications. Then, it adjusts the medication plan based on personalized treatment parameters, including specific drug selection, dosage calculation, medication timing, and treatment duration. Finally, it generates a structured treatment recommendation report, which details the specific diagnostic conclusions, treatment plan content, and medication precautions.

[0078] In one embodiment, refer to Figure 2 In step S200, similarity is calculated and diagnostic information exceeding a preset threshold is filtered to obtain a preliminary diagnostic and treatment plan, which specifically includes the following steps:

[0079] S210. Calculate the cosine similarity between the symptom feature vector and the diagnosis and treatment node vector in the knowledge graph to obtain a similarity score sequence.

[0080] In this embodiment, the diagnosis and treatment node vector refers to the standardized feature vector corresponding to each disease diagnosis and treatment plan in the knowledge graph, which includes numerical representations of four dimensions: symptom type, symptom severity, location of occurrence, and duration; cosine similarity refers to the method of measuring the similarity between vectors by calculating the cosine value of the angle between two vectors, with a value range of 0-1; the similarity score sequence refers to the ordered sequence of similarity values ​​calculated between the input vector and all diagnosis and treatment node vectors.

[0081] Specifically, a similarity calculation module is established, using the standard cosine similarity formula. First, the input symptom feature vector is compared one by one with the multiple diagnosis and treatment node vectors in the knowledge graph, and the dot product of the two vectors is calculated and divided by the product of the vector magnitudes.

[0082] S220. Based on the similarity score sequence, threshold screening is performed to select diagnosis and treatment nodes with similarity greater than the preset threshold, thereby obtaining a set of candidate diagnosis and treatment plans.

[0083] In this embodiment, all schemes with a similarity greater than a preset threshold are included in the candidate set, while schemes with a similarity between 0.6 and 0.8 are retained as alternative schemes. The maximum number of candidate schemes is set to five to avoid an excessive number of candidates. The system sorts the selected candidate schemes in descending order of similarity, forming a candidate treatment scheme set.

[0084] Furthermore, this application pre-establishes a preset threshold library. This library is based on extensive clinical data statistics and expert experience, and refers to a dataset that sets personalized similarity thresholds for different diagnostic and treatment nodes. The diagnostic and treatment node features include quantitative indicators across three dimensions: symptom specificity, symptom variability, and diagnostic complexity. The personalized thresholds refer to adaptively adjusted screening criteria based on node features, with a value range of 0.6-0.95. Specifically, a threshold adaptive adjustment system is established, including a node feature scoring table, a threshold benchmark table, and an adjustment coefficient table. The node feature scoring table records the quantitative scores of each diagnostic and treatment node across the three dimensions; the threshold benchmark table sets benchmark thresholds for different feature combinations; and the adjustment coefficient table defines the weight of the feature scores on the thresholds. Differentiated thresholds are set for different types of diseases: common illnesses like the common cold and diarrhea, with less specific symptom combinations, are assigned a lower threshold of 0.65; diseases like systemic lupus erythematosus, with diverse and fluctuating symptoms, are assigned a medium threshold of 0.75; diseases like rheumatoid arthritis, which require multiple diagnostic criteria, are assigned a higher threshold of 0.85; and rare or critical illnesses like myasthenia gravis, due to their highly specific symptoms, are assigned a maximum threshold of 0.95. When performing similarity screening, the system first queries the preset threshold of the target diagnostic node, and then uses this threshold as the screening criterion for candidate solutions. This differentiated threshold setting can more accurately identify different types of diseases and improve diagnostic accuracy.

[0085] S230. Based on the set of candidate treatment plans, the plans are integrated by weighting each candidate plan according to the weight of the similarity score to obtain a preliminary treatment plan.

[0086] In this embodiment, scheme fusion refers to the process of integrating multiple candidate treatment schemes into a comprehensive scheme; weight refers to the importance coefficient of each candidate scheme calculated based on the similarity score, which is obtained by dividing the similarity value by the sum of the similarities of all candidate schemes; weighted integration refers to the reasonable combination of treatment measures in each candidate scheme according to the weight to generate a comprehensive treatment scheme.

[0087] In one embodiment, refer to Figure 3 In step S230, the candidate treatment plan set includes a subset of traditional Chinese medicine treatment plans and a subset of Western medicine treatment plans. The plans are then fused, specifically including the following steps:

[0088] S231. Perform prescription analysis on the prescriptions in the subset of TCM treatment plans, extract the dosage and ratio information of the main and auxiliary drugs, and obtain the compatibility characteristics of TCM.

[0089] In this embodiment, the compatibility characteristics of traditional Chinese medicine refer to the combination relationship and dosage ratio of each drug in the prescription.

[0090] Specifically, a prescription analysis system is established, comprising three basic data tables: a drug attribute table, a compatibility table, and a dosage standard table. The system first identifies the principal, assistant, adjuvant, and guide herbs in the prescription, extracting information on the principal and auxiliary herbs; then it analyzes the dosage of each herb and calculates the ratio of principal to auxiliary herbs; finally, based on the properties of the herbs and their compatibility relationships, it forms a complete description of the compatibility characteristics. For example, for a prescription for treating a cold, the system identifies honeysuckle and forsythia as the principal herbs, 15 grams each, and platycodon and licorice as auxiliary herbs, 6 grams and 3 grams respectively, resulting in a principal-auxiliary ratio of 2.5:1.

[0091] S232. Perform drug analysis on prescriptions in the subset of Western medicine treatment plans, extract drug targets and dosage information, and obtain the characteristics of Western medicine use.

[0092] In this embodiment, the characteristics of Western medicine use refer to the combination of the drug's action characteristics and specific usage requirements.

[0093] Specifically, a Western medicine analysis module is established, containing three core data tables: a drug information table, a dosing regimen table, and a dosage conversion table. The system analyzes each drug in the prescription, extracts its target information, and records the specific dosage and medication requirements.

[0094] S233. Based on a pre-set database of rules for the combined use of traditional Chinese medicine and Western medicine, conduct compatibility checks, analyze the interaction between the compatibility characteristics of traditional Chinese medicine and the characteristics of Western medicine, and obtain a coordinated medication plan.

[0095] In this embodiment, the rule base for combined use of traditional Chinese and Western medicines refers to a structured database that records the interaction patterns of traditional Chinese and Western medicines; compatibility testing refers to analyzing the possible effects when traditional Chinese and Western medicines are used in combination; interactions include four types: enhanced efficacy, weakened efficacy, enhanced toxicity, and metabolic effects; and the drug coordination plan refers to a safe and effective combined drug use plan determined after compatibility testing.

[0096] S234. Based on the medication coordination plan, integrate the plans, adjust the ratio and dosage of Chinese and Western medicines, and generate a preliminary treatment plan.

[0097] Specifically, the system first calculates the adjusted medication dosage and determines the medication schedule based on the recommendations in the medication coordination plan. Then, it generates a structured treatment plan, including specific medication orders, medication requirements, and precautions.

[0098] In one embodiment, refer to Figure 4 In step S300, based on a preset medication contraindication rule base, personalized treatment parameters for the patient are calculated, specifically including the following steps:

[0099] S310. Based on a preset medication contraindication rule library, extract standard medication parameters and use the standard medication parameters as benchmark treatment parameters.

[0100] In this embodiment, standard medication parameters refer to the set of baseline parameters for drug use, including standard dosage range, dosing frequency, timing of dosing, and duration of treatment.

[0101] S320. Read individual patient information to obtain tolerance and corresponding safety threshold parameters for different drug treatment effects.

[0102] In this embodiment, drug tolerance refers to the patient's ability to tolerate the drug (1-10 points); the safety threshold parameter includes three specific values: maximum tolerated dose, adverse reaction warning value, and monitoring indicator range.

[0103] Specifically, the system analyzes patient information, calculates tolerance scores for different types of drugs, and determines corresponding safe medication thresholds. For example, for an elderly patient, the system might lower their tolerance score for certain drugs and correspondingly reduce the safety threshold parameters.

[0104] S330. Based on the patient's historical medication records, determine the efficacy characteristics and dosage adjustment range of the corresponding drug, and combine the tolerance of different drug therapeutic effects and the corresponding safety threshold parameters to determine the target dosage and dynamic control parameters of medication.

[0105] Among them, the dynamic control parameters for medication include the dosing range parameters based on different therapeutic effects; efficacy characteristics refer to the quantitative indicators of drug treatment effects; and the dose adjustment range refers to the safe dose range determined based on individual patient responses, including three specific values: minimum effective dose, optimal therapeutic dose, and maximum tolerated dose.

[0106] Specifically, the system first analyzes the efficacy data in historical medication records to extract the characteristics of the drug's action in the patient's body; then, it combines the tolerance score to determine a personalized dose adjustment range.

[0107] S340. Based on the target dosage and dynamic drug control parameters, obtain the patient's personalized treatment parameters.

[0108] In this embodiment, the target dosage refers to the optimal treatment dose determined based on the individual characteristics of the patient; the dynamic control parameters for medication include the initial dose, the rate of increase, the maintenance dose, and the adjustment criteria.

[0109] Specifically, the system generates a detailed medication regimen based on the target dosage and dynamic control parameters.

[0110] In one embodiment, refer to Figure 5 In step S400, based on the preliminary treatment plan and the patient's personalized treatment parameters, the plan is optimized to generate the final integrated traditional Chinese and Western medicine treatment recommendations, which specifically include the following steps:

[0111] S410. Based on the preliminary treatment plan, analyze the plan, extract the traditional Chinese medicine prescription and Western medicine medication information respectively, and obtain the components of the treatment plan.

[0112] Specifically, the system first categorizes prescriptions into traditional Chinese medicine and Western medicine, and extracts their respective medication information; then, it reorganizes the extracted information according to a standard format to form a structured list of components.

[0113] S420. Based on the patient's personalized treatment parameters, adjust the dosage and optimize the drug dosage in the components of the treatment plan to obtain the adjusted medication plan.

[0114] In this embodiment, the system pre-establishes a dose conversion table (recording dose conversion coefficients for different populations), an adjustment rule table (recording specific standards for dose adjustment), a safety range table (recording the safe range for medication use), and a verification standard table (recording the criteria for determining the rationality of the dose). The system first calculates the specific dosage of each drug based on the patient's individual parameters; then, it verifies the safety of the calculation results to ensure they are within a reasonable range. For example, for an elderly hypertensive patient, the system will adjust the standard dose of amlodipine from 5 mg to 2.5 mg based on their age and renal function, and then gradually adjust it based on the effect after one week of observation.

[0115] S430. Based on the dosing range parameter and drug interaction limit threshold in the dynamic control parameters of medication, perform constraint testing to verify whether the adjusted medication regimen meets the constraint conditions and obtain the verification results.

[0116] Specifically, the system first checks whether the dosage of each drug is within the safe range; then it verifies whether the dosing schedule is reasonable; and finally, it analyzes whether drug interactions exceed the limit threshold. For example, when the system detects that the time interval between taking a certain antihypertensive traditional Chinese medicine and an antihypertensive Western medicine is less than the recommended value, it will mark this item as an item to be adjusted and provide specific adjustment suggestions.

[0117] Furthermore, firstly, the items to be adjusted are graded using a priority scoring table, which includes three dimensions: safety risk score, efficacy impact score, and adjustment difficulty score. Items with high risk, significant impact, and low adjustment difficulty are prioritized. Then, alternative plans are generated based on the adjustment strategy table, including a proportionally decreasing dose adjustment plan, a staggered treatment plan at fixed time intervals, a segmented treatment plan with a span of days, and several alternative drug plans selected from similar drugs. Next, the plan is validated using a validation index table, which sets three validation dimensions: safety index for adverse reaction incidence, effectiveness index for treatment goal achievement rate, and compliance index for patient cooperation score. Finally, the optimal plan is determined using a plan evaluation table.

[0118] S440. Based on the validation results, generate a treatment plan, integrate and adjust the medication plan, and generate the final integrated traditional Chinese and Western medicine treatment recommendations.

[0119] Specifically, the system first integrates validated medication regimens; then generates detailed medication instructions and precautions; and finally develops corresponding monitoring and follow-up plans.

[0120] In one embodiment, refer to Figure 6 In step S340, the dynamic drug control parameters include thresholds for limiting drug interactions. After obtaining the patient's personalized treatment parameters based on the target dosage and the dynamic drug control parameters, the following steps are also included:

[0121] S610. Based on the different therapeutic effects of drugs, generate several dose control units within each administration range.

[0122] In this embodiment, the dose control unit refers to the smallest dose adjustment unit within the administration range, which includes the base dose value, adjustment step size, time span, and monitoring indicators; the administration range refers to the safe dose interval corresponding to the therapeutic effect of the drug, which is determined by the minimum effective dose and the maximum tolerated dose.

[0123] S620. In the preset dynamic control drug delivery model, each dose control unit is used as a basic unit. Target drug delivery values ​​are set for all therapeutic drugs within the dose control unit, and the sum of the effects of all therapeutic drugs within the dose control unit is set to equal the preset therapeutic effect threshold.

[0124] In this embodiment, the dynamic control dosing model refers to a mathematical model for controlling the combined use of multiple drugs; the target dosing value refers to the specific dosage of the drug used in each dose control unit; the effect intensity refers to the degree to which the drug exerts its therapeutic effect; and the preset treatment effect threshold refers to the total score of the expected treatment effect.

[0125] Specifically, the system first calculates the efficacy of different doses of each drug, and then uses an optimization algorithm to determine the optimal dose combination that can achieve the preset therapeutic effect threshold.

[0126] S630. Based on the identified current drug combination and current dosing regimen, obtain the corresponding drug interaction limit threshold; using each dose control unit as the basic unit, control the interaction intensity of all therapeutic drugs in each dose control unit to be less than or equal to the corresponding limit threshold.

[0127] In this embodiment, the drug interaction limitation threshold refers to the maximum acceptable interaction strength of the drug combination; the interaction strength refers to the degree of mutual influence between drugs; the current drug combination refers to the set of drugs used together within the same time period; and the current dosing regimen refers to the specific dosage and timing of these drugs.

[0128] Specifically, the system first identifies all currently used drug combinations and queries their interaction types and strength scores. Then, based on individual patient characteristics, it determines specific limit thresholds for each combination. For example, for a combination containing an antihypertensive and a hypoglycemic agent, the system finds the interaction strength to be 3 points. Considering the patient's age, the limit threshold is set to 4 points, requiring monitoring of blood pressure and blood glucose. Within each dose control unit, the system ensures that the drug interaction strength remains below this limit threshold, adjusting the dosing time or dosage as necessary. When the interaction strength within a dose control unit exceeds the limit threshold, the system automatically generates adjustment suggestions: adjusting the dosing time, reducing the dosage of certain drugs, or changing the route of administration. For example, when the interaction strength between an antihypertensive and a hypoglycemic agent approaches the limit threshold, the system suggests extending the dosing interval between the two drugs to 4 hours and performing dynamic monitoring of blood pressure and blood glucose after administration.

[0129] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0130] Secondly, this application provides an intelligent decision-making and diagnosis system that integrates traditional Chinese and Western medicine. The intelligent decision-making and diagnosis system that integrates traditional Chinese and Western medicine of this application will be described below in conjunction with the above-mentioned intelligent decision-making and diagnosis method that integrates traditional Chinese and Western medicine.

[0131] Reference Figure 7 A smart decision-making and treatment system integrating traditional Chinese and Western medicine includes:

[0132] The symptom input module is used to receive patient symptom input, encode the symptoms, and calculate the symptom feature vector.

[0133] The treatment plan matching module is used to perform vector matching in a preset integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph based on symptom feature vectors, calculate similarity and filter diagnosis and treatment information that exceeds a preset threshold to obtain a preliminary diagnosis and treatment plan.

[0134] The parameter calculation module is used to read individual patient information, perform feature analysis, and calculate personalized treatment parameters for patients based on a preset medication contraindication rule library.

[0135] The treatment plan optimization module is used to optimize the treatment plan based on the initial diagnosis and treatment plan and the patient's personalized treatment parameters, and generate the final integrated traditional Chinese and Western medicine treatment recommendations.

[0136] In one embodiment, the scheme matching module includes:

[0137] The similarity calculation unit is used to calculate the cosine similarity between the symptom feature vector and the diagnosis and treatment node vector in the knowledge graph to obtain a similarity score sequence.

[0138] The filtering unit is used to perform threshold filtering based on the similarity score sequence, filter out the diagnosis and treatment nodes with similarity greater than the preset threshold, and obtain a set of candidate diagnosis and treatment plans;

[0139] The scheme fusion unit is used to fuse schemes based on the set of candidate treatment schemes. It integrates each candidate scheme by weighting it according to the similarity score to obtain a preliminary treatment scheme.

[0140] In one embodiment, this application provides an electronic device, which may be a server, and its internal structure diagram may be as follows: Figure 8As shown, this electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements an intelligent decision-making and diagnostic method integrating traditional Chinese and Western medicine.

[0141] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0142] In one embodiment, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0143] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0144] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A smart decision-making and treatment method integrating traditional Chinese and Western medicine, characterized in that, Includes the following steps: Receive patient symptom input, encode the symptoms, and calculate the symptom feature vector; Based on the symptom feature vector, vector matching is performed in a preset integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph to calculate the similarity and filter out diagnosis and treatment information that exceeds a preset threshold to obtain a preliminary diagnosis and treatment plan. Read individual patient information, perform feature analysis, and calculate personalized treatment parameters for the patient based on a pre-set medication contraindication rule base; Based on the preliminary treatment plan and the patient's personalized treatment parameters, the plan is optimized to generate the final integrated traditional Chinese and Western medicine treatment recommendations. The process of calculating similarity and filtering out medical information that exceeds a preset threshold to obtain a preliminary treatment plan includes the following steps: The cosine similarity between the symptom feature vector and the diagnosis and treatment node vector in the knowledge graph is calculated to obtain a similarity score sequence. Based on the similarity score sequence, threshold filtering is performed to select diagnostic and treatment nodes with similarity greater than a preset threshold, thereby obtaining a set of candidate diagnostic and treatment plans. Based on the set of candidate treatment plans, the plans are fused together. The candidate plans are weighted and integrated according to the weight of the similarity scores to obtain a preliminary treatment plan. The candidate treatment plan set includes a subset of traditional Chinese medicine treatment plans and a subset of Western medicine treatment plans. The plan is then fused, specifically including the following steps: For the prescriptions in the subset of the TCM treatment plan, formula analysis is performed to extract the dosage and ratio information of the main and auxiliary drugs, and the compatibility characteristics of TCM are obtained. Drug analysis is performed on the prescriptions in the subset of Western medicine treatment protocols to extract drug targets and dosage information, thereby obtaining the characteristics of Western medicine use. Based on a pre-defined database of rules for combined use of traditional Chinese and Western medicines, compatibility checks are performed, and the interaction between the compatibility characteristics of traditional Chinese medicines and the usage characteristics of Western medicines is analyzed to obtain a coordinated medication plan. The database of rules for combined use of traditional Chinese and Western medicines refers to a structured database that records the interaction patterns between traditional Chinese and Western medicines. The interaction includes four types: enhanced efficacy, weakened efficacy, enhanced toxicity, and metabolic effects. Based on the aforementioned medication coordination plan, the plans are integrated, the ratio and dosage of Chinese and Western medicines are adjusted, and a preliminary treatment plan is generated.

2. The intelligent decision-making and treatment method combining traditional Chinese and Western medicine according to claim 1, characterized in that, Based on a pre-defined database of medication contraindications, personalized treatment parameters for each patient are calculated, including the following steps: Based on a pre-defined database of contraindications for medication, standard medication parameters are extracted and used as baseline treatment parameters. Read individual patient information to obtain tolerance and corresponding safety threshold parameters for different drug treatments; Based on the patient's historical medication records, the therapeutic characteristics and dosage adjustment range of the corresponding drugs are determined. Combined with the tolerability of the different therapeutic effects of the drugs and the corresponding safety threshold parameters, the target dosage and dynamic drug control parameters are determined. The dynamic drug control parameters include dosage range parameters based on different therapeutic effects. Based on the target dosage and the dynamic drug control parameters, personalized treatment parameters for the patient are obtained.

3. The intelligent decision-making and diagnosis method combining traditional Chinese and Western medicine according to claim 2, characterized in that, The dynamic drug control parameters include thresholds limiting different drug interactions; after obtaining the patient's personalized treatment parameters based on the target dosage and the dynamic drug control parameters, the following steps are also included: Based on the different therapeutic effects of drugs, several dose control units are generated within each administration range; In the preset dynamic control drug delivery model, each dose control unit is used as a basic unit, and a target dosing value is set for all therapeutic drugs in the dose control unit. The sum of the effects of all therapeutic drugs in the dose control unit is set to be equal to the preset therapeutic effect threshold. Based on the identified current drug combination and current dosing regimen, the corresponding drug interaction limitation threshold is obtained; taking each dose control unit as a basic unit, the interaction intensity of all therapeutic drugs in each dose control unit is controlled to be less than or equal to the corresponding limitation threshold.

4. The intelligent decision-making and diagnosis method combining traditional Chinese and Western medicine according to claim 3, characterized in that, Based on the preliminary treatment plan and the patient's personalized treatment parameters, the plan is optimized to generate a final integrated traditional Chinese and Western medicine treatment recommendation, which includes the following steps: Based on the preliminary treatment plan, the plan was analyzed, and the prescriptions for traditional Chinese medicine and the medication information for Western medicine were extracted to obtain the components of the treatment plan. Based on the patient's personalized treatment parameters, the dosage is adjusted, and the medication dosage in the components of the treatment plan is optimized to obtain the adjusted medication plan. Based on the dosing range parameter and drug interaction limitation threshold in the medication dynamic control parameters, a constraint test is performed to verify whether the adjusted medication regimen meets the constraint conditions, and the verification results are obtained. Based on the verification results, a treatment plan is generated, and the adjusted medication plan is integrated to generate the final integrated traditional Chinese and Western medicine treatment recommendations.

5. An intelligent decision-making and diagnosis system integrating traditional Chinese and Western medicine, characterized in that, The intelligent decision-making diagnosis and treatment method combining traditional Chinese and Western medicine, as described in any one of claims 1-4, includes: The symptom input module is used to receive patient symptom input, encode the symptoms, and calculate the symptom feature vector. The treatment plan matching module is used to perform vector matching in a preset integrated traditional Chinese and Western medicine diagnosis and treatment knowledge graph based on the symptom feature vector, calculate the similarity and filter out diagnosis and treatment information that exceeds a preset threshold to obtain a preliminary diagnosis and treatment plan. The parameter calculation module is used to read individual patient information, perform feature analysis, and calculate personalized treatment parameters for patients based on a preset medication contraindication rule library. The treatment plan optimization module is used to optimize the treatment plan based on the preliminary treatment plan and the patient's personalized treatment parameters, and generate the final integrated traditional Chinese and Western medicine treatment recommendations.

6. The intelligent decision-making and diagnosis system integrating traditional Chinese and Western medicine according to claim 5, characterized in that, The scheme matching module includes: The similarity calculation unit is used to perform cosine similarity calculation between the symptom feature vector and the diagnosis and treatment node vector in the knowledge graph to obtain a similarity score sequence. The filtering unit is used to perform threshold filtering based on the similarity score sequence, filter out the diagnosis and treatment nodes with similarity greater than a preset threshold, and obtain a set of candidate diagnosis and treatment plans. The scheme fusion unit is used to perform scheme fusion based on the set of candidate treatment schemes, and to integrate each candidate scheme by weighting according to the weight of the similarity score to obtain a preliminary treatment scheme.

7. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the intelligent decision-making and diagnosis method combining traditional Chinese and Western medicine as described in any one of claims 1-4.

Images