Western medicine pharmaceutical ingredient data intelligent storage management method

Abstract

The invention discloses an intelligent storage management method for western medicine pharmaceutical ingredient data, and relates to the technical field of medical information, the method comprises the following steps: collecting medicine ingredient data through an API interface, carrying out semantic analysis and encrypted storage by using a natural language processing technology, and constructing a medicine self-recommendation platform; personalized medicine recommendation is provided for patients through electronic medical record analysis and retrieval of medicine component label attributes. Through calculation of a drug recommendation validity index ytx, drug recommendation validity can be preliminarily evaluated, gene compatibility analysis is further performed in combination with gene drug efficacy data of a patient, and after a gene compatibility analysis instruction is received, a gene compatibility index jsp is calculated. And finally, through calculation of a comprehensive gene adaptation index zht, evaluating the matching between the drug and the gene of the patient, and optimizing a drug recommendation scheme. According to the method, the accuracy and individuation level of drug recommendation can be effectively improved, side effects and treatment risks of drugs are reduced, and technical support is provided for individualized medical treatment.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and particularly to an intelligent storage and management method for western medicine pharmaceutical ingredient data. Background Art

[0002] With the rapid development of technology, the wide application of information technology, especially big data, artificial intelligence and cloud computing, has profoundly affected the development of all walks of life. Especially in the field of medical health, personalized medicine and precision treatment have gradually become important trends. With the continuous accumulation of patient genomics, pathological data and drug ingredients, the traditional drug treatment model can no longer fully meet the needs of complex etiologies and individual differences. Therefore, how to achieve the precise adaptation of drugs and genes through intelligent data storage and analysis technology has become an important direction to improve treatment effects, reduce side effects and lower medical costs. At the same time, the society's demand for the security, privacy protection of medical data and precision medicine is constantly increasing, which has promoted the rapid development of drug recommendation and gene matching technology based on intelligence and informatization.

[0003] Despite certain progress in the research and application of drug-gene adaptation, the existing systems still have some deficiencies. Existing drug self-recommendation platforms usually rely on fixed disease classifications and drug indication information, which may lead to a reduction in the accuracy of individualized treatment, especially when facing complex etiologies and different patient gene backgrounds. In addition, current drug recommendation algorithms mostly evaluate the efficacy and adaptability of drugs based on the ingredient correlation degree of drugs and the drug indication matching index, but these methods often ignore key factors such as the patient's gene background and metabolic pathways. Furthermore, the matching degree of existing drug self-recommendation platforms does not always accurately reflect the individual differences of patients, resulting in a relatively high failure rate of drug adaptation. The evaluation mechanism of the drug recommendation validity index is also often not easy to adapt to changes in different clinical environments. Especially when facing complex multiple drug use and drug interactions, the existing systems still lack sufficient flexibility and precision. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides an intelligent storage and management method for western medicine pharmaceutical ingredient data, which solves the problems in the above background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent storage and management method for western medicine pharmaceutical ingredient data, comprising the following steps:

[0006] S1. Build an intelligent data storage repository based on the cloud platform, and then collect the drug ingredient data of western medicine from the western medicine data source through the API dedicated interface, perform semantic parsing and tagging, and then encrypt and store it in the intelligent data storage repository;

[0007] S2. The cloud platform constructs a drug self-recommendation platform, locates the drugs of all potential therapeutic ingredients in the intelligent data repository through the electronic medical records of patients, decrypts them temporarily, retrieves them using the conditional combination of label attributes after decryption, and extracts the information data of the drugs.

[0008] S3. Based on the extracted drug information data, conduct drug ingredient correlation analysis and drug indication analysis, and perform summary calculations based on the analysis results to obtain the drug recommendation validity index ytx for preliminary drug matching evaluation. If the drug recommendation validity index ytx exceeds the drug adaptation threshold A, then perform gene compatibility analysis.

[0009] S4. After receiving the gene compatibility analysis instruction, obtain the gene efficacy data of the patient according to the multi-dimensional data analysis technology, and based on the gene efficacy data, analyze the compliance of the patient's medication needs and physiological status to obtain the gene adaptability index jsp.

[0010] S5. Perform comprehensive calculations on the drug recommendation validity index ytx and the gene adaptability index jsp to obtain the comprehensive gene adaptability index zht for medication and gene matching evaluation.

[0011] Preferably, S1 includes S11, S12, and S13.

[0012] S11. The cloud platform collects the drug ingredient data of each western medicine from the western medicine data source through the API dedicated interface, cleans and standardizes it, and then uses the natural language processing NLP technology to perform semantic parsing on the drug description.

[0013] The western medicine data source includes drug databases, literature resources, clinical trials, drug regulatory agencies, and pharmaceutical factories.

[0014] S12. Use the cloud platform to construct an intelligent data repository, structurally process the drug ingredient data, identify the drug ingredient attributes, including drug categories, drug ingredients, and drug efficacy, judge the sensitivity level of the drugs according to the ingredient attributes, and then use the metadata index to label each drug based on the drug ingredient attributes and sensitivity level, and store the labeled drug ingredient data in the intelligent data repository.

[0015] The sensitivity levels are divided into anesthetic drugs, state-controlled drugs, and confidential drugs.

[0016] S13. The intelligent data repository enables server-side encryption, encrypts each drug data in the stored drug ingredient data with a corresponding unique encryption key, and uniformly manages it through the key management service KMS of the cloud platform.

[0017] Preferably, S2 includes S21 and S22.

[0018] S21. After the doctor diagnoses the cause of the patient's illness and generates an electronic medical record, a drug self-recommendation platform is constructed through the cloud platform. The drug self-recommendation platform uses medical natural language processing technology to segment the electronic medical record and identify named entities, map them to standard cause-of-disease codes, establish a ternary map of diseases, symptoms, and drug ingredients, and locate all drugs with potential therapeutic ingredients corresponding to the cause of the disease in the intelligent data repository;

[0019] S22. The drug self-recommendation platform temporarily decrypts the located drugs with potential therapeutic ingredients through a temporary key. The drug self-recommendation platform uses conditional combinations of tag attributes carried by each drug ingredient for retrieval, extracts drugs suitable for specific pathological characteristics and individual conditions, recommends drugs to the patient, and extracts information data of the drugs;

[0020] The information data includes drug characteristic dimensions and drug indications.

[0021] Preferably, S3 includes S31 and S32;

[0022] S31. After dimensionless processing of the drug ingredient correlation analysis and drug indication analysis based on the extracted drug information data, the drug ingredient correlation index ywg and the drug indication matching index spp are respectively obtained by summarizing and calculating;

[0023] The drug ingredient correlation index ywg is obtained through the following formula;

[0024] ;

[0025] In the formula, ywg ij represents the correlation between drug i and drug j, n represents the number of characteristic dimensions, w k represents the weight of the kth feature, and respectively represent the performance values of drug i and drug j on the kth characteristic dimension, i and j represent index symbols, i ∈ {1, 2,..., N}, j ∈ {1, 2,..., N}, and are indexed in the set of the total number of drugs N extracted;

[0026] The drug indication matching index spp is obtained through the following formula;

[0027] ;

[0028] In the formula, spp ij represents the drug indication matching index between drug i and drug j, represents the number of intersections of the indications of drug i and drug j, represents the number of unions of the indications of drug i and drug j.

[0029] Preferably, S32 is used to preliminarily judge the comprehensive suitability of the recommended drug and the patient's cause of disease, specifically including S321 and S322;

[0030] S321, after dimensionless processing and summary calculation based on the obtained drug component correlation index ywg and drug indication matching index spp, obtains the drug recommendation validity index ytx, which quantifies the comprehensive suitability of the recommended drug and the patient's cause of disease;

[0031] ;

[0032] In the formula, cy i and cy j respectively represent the metabolic utility factors of drug i and drug j under the specific conditions of the patient, represents the difference factor of the metabolic utility of drug i and drug j under the specific conditions of the patient.

[0033] Preferably, S322, based on the drug recommendation validity indexes ytx of all historical samples with the same cause of disease, calculates the mean u and standard deviation , and based on the mean u and standard deviation presets the drug adaptation threshold A, specifically: , where, represents the confidence factor, which is used to adjust the sensitivity of the threshold, and then conducts a preliminary drug matching evaluation with the obtained drug recommendation validity index ytx. The specific evaluation scheme is as follows;

[0034] When the drug recommendation validity index ytx > the drug adaptation threshold A, the drug matching is successful, and at this time, gene compatibility analysis is triggered;

[0035] When the drug recommendation validity index ytx ≤ the drug adaptation threshold A, the drug matching fails, and at this time, the drug self-recommendation platform automatically performs iterative optimization matching until the matching is successful.

[0036] Preferably, S4 is used to perform gene compatibility analysis when the preliminary drug matching evaluation is successful for drug matching, specifically including S41 and S42;

[0037] S41, the gene compatibility analysis marks the successfully matched drug as drug X, and conducts a trial on the patient, and then obtains the gene pharmacodynamic data of the patient based on the multi-dimensional data analysis technology;

[0038] The multi-dimensional data analysis technology includes whole-genome sequencing technology, electronic medical record system, and metabolic pathway-related gene expression analysis;

[0039] Using whole-genome sequencing technology to perform drug metabolism detection on patients, matching the detection results with the metabolic pathway matching table in the intelligent data repository, and obtaining the gene adaptation factor sp by calculating the ratio of the matching channels to the total channels;

[0040] Extract the changes in physiological indicators after medication from the patient's electronic medical record system to obtain the physiological condition compatibility jr, and then compare it with the physiological indicators before medication to obtain the drug efficacy response yl;

[0041] Based on the analysis of the gene expression related to the patient's metabolic pathway, use the intelligent data repository to establish a drug metabolic pathway mapping diagram, analyze the matching of pathway nodes through graph theory, and compare the number of complete key enzymes expressed with the total number of enzymes required for the drug to obtain the metabolic pathway matching value dx;

[0042] The gene drug efficacy data includes the gene adaptation factor sp, the physiological condition compatibility jr, the drug efficacy response yl, and the metabolic pathway matching value dx.

[0043] Preferably, in S42, after performing dimensionless processing on the obtained gene drug efficacy data and summarizing the calculations, the gene adaptation index jsp is obtained;

[0044] The gene adaptation index jsp is obtained through the following formula to analyze the compliance between the patient's medication needs and physiological status;

[0045] ;

[0046] In the formula, e represents the exponential function, m represents the total number of gene loci, p represents the gene locus number, represents the adaptation factor of drug X at the pth gene locus, and yl X represents the drug efficacy response of drug X to the patient, represents the metabolic pathway matching value of drug X on the pth gene, and jr X represents the physiological condition compatibility of drug X to the patient.

[0047] Preferably, S5 includes S51 and S52;

[0048] S51, based on the drug recommendation validity index ytx and the gene adaptation index jsp, perform comprehensive calculations to obtain the comprehensive gene adaptation index zht, which quantifies the adaptation level of the drug required for gene matching;

[0049] The comprehensive gene adaptation index zht is obtained through the following formula;

[0050] ;

[0051] In the formula, β represents the weight factor.

[0052] Preferably, in S52, a preset gene adaptation threshold Z is set based on the drug neighborhood gene adaptation standard, and the obtained comprehensive gene adaptation index zht is used to evaluate the drug-gene matching. The specific evaluation scheme is as follows;

[0053] When the comprehensive gene adaptation index zht > the gene adaptation threshold Z, it indicates that there is no hidden danger at the drug-gene level and it meets the patient's medication needs;

[0054] When the comprehensive gene adaptation index zht ≤ the gene adaptation threshold Z, it indicates that there is a hidden danger at the drug-gene level and it does not meet the patient's medication needs. At this time, the drug self-recommendation platform performs iterative optimization matching through S3 until the matching is successful.

[0055] The present invention provides an intelligent storage management method for western medicine pharmaceutical ingredient data. It has the following beneficial effects:

[0056] (1) This method constructs an intelligent data storage repository through a cloud platform, collects the drug ingredient data of western medicine using an API interface, performs semantic parsing, and classifies the drugs using tagged management, and finally encrypts and stores them in the intelligent data storage repository. This process not only ensures the security of drug data, but also improves the organization and retrieval efficiency of drug data by comprehensively considering uncommon parameters such as drug ingredient attributes, pharmacodynamic mechanisms, and sensitivity levels. This process lays a solid foundation for subsequent personalized drug recommendations.

[0057] (2) The drug self-recommendation platform of this method is based on the electronic medical record and the patient's cause of disease, combines the tagged attributes of drugs, and extracts and matches drug information through medical natural language processing technology and multi-dimensional data analysis technology. In this process, the similarity and indications of drugs are quantitatively analyzed by calculating the drug ingredient correlation index ywg and the drug indication matching index spp. In the drug recommendation validity evaluation, the drug recommendation validity index ytx is used to initially measure the adaptability of the drug to the patient's cause of disease, and the historical data is analyzed by statistical methods to set the drug adaptation threshold A. When the drug recommendation validity index ytx exceeds the drug adaptation threshold A, it is considered that the recommended drug is successfully matched, and the gene compatibility analysis is started.

[0058] (3)In the gene compatibility analysis, this method obtains the gene pharmacodynamic data of patients through multi-dimensional data analysis technology, aggregates and calculates to obtain the gene adaptability index jsp, and combines it with the drug recommendation validity index ytx for comprehensive calculation to obtain the comprehensive recommendation validity index zht, further optimizing the drug recommendation process. Through the comparison and evaluation with the gene adaptation threshold Z, it ensures the matching of drugs and genes, ensures the safety and treatment effect of patients' medications, and ultimately achieves the goal of accurate personalized drug recommendation and optimized patient treatment plans. By integrating multi-dimensional data analysis and intelligent storage technology, this method realizes the intelligent storage of western medicine pharmaceutical ingredients and personalized drug recommendation for patients, solves the problem of insufficient drug matching accuracy, provides a more scientific and personalized treatment plan, and provides higher-quality medical services for patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic diagram of the steps of a method for intelligent storage and management of western medicine pharmaceutical ingredient data according to the present invention;

[0060] Figure 2 It is a linear schematic diagram of the drug recommendation validity relationship according to the present invention;

[0061] Figure 3 It is a schematic diagram of the technical relationship of a method for intelligent storage and management of western medicine pharmaceutical ingredient data according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0063] Embodiment 1

[0064] Please refer to Figure 1 and Figure 3 , the present invention provides a method for intelligent storage and management of western medicine pharmaceutical ingredient data. To achieve the above objectives, the present invention is implemented through the following technical solutions: including the following steps:

[0065] S1. Build an intelligent data storage repository based on the cloud platform, and then collect the drug ingredient data of western medicine from the western medicine data source through the API dedicated interface, perform semantic parsing and tagging, and then encrypt and store it in the intelligent data storage repository;

[0066] S2. The cloud platform constructs a drug self-recommendation platform, locates the drugs of all potential therapeutic ingredients in the intelligent data repository through the electronic medical records of patients, decrypts them temporarily, retrieves them using conditional combinations according to label attributes, and extracts the information data of the drugs;

[0067] S3. Conduct drug component correlation analysis and drug indication analysis based on the extracted drug information data, and perform summary calculations based on the analysis results to obtain the drug recommendation validity index ytx for preliminary drug matching evaluation. If the drug recommendation validity index ytx exceeds the drug adaptation threshold A, then perform gene compatibility analysis;

[0068] S4. After receiving the gene compatibility analysis instruction, obtain the gene pharmacodynamic data of the patient based on the multi-dimensional data analysis technology, and analyze the compliance of the patient's medication needs and physiological state based on the gene pharmacodynamic data to obtain the gene adaptability index jsp;

[0069] S5. Perform comprehensive calculations on the drug recommendation validity index ytx and the gene adaptability index jsp to obtain the comprehensive gene adaptation index zht for evaluating the matching of medication and genes.

[0070] In this embodiment, an intelligent data repository is built through a cloud platform. By combining with the API interface, drug ingredient data is collected from multiple Western medicine data sources, and then subjected to semantic parsing and encrypted storage, ensuring the comprehensiveness, security, and real-time nature of drug data. In addition, the drug self-recommendation platform combines with the patient's electronic medical record for precise matching, not only improving the efficiency of drug recommendation but also ensuring the accuracy of the recommended drugs. Through preliminary analysis of drug ingredients and indications, and introducing the drug recommendation validity index ytx for preliminary drug matching evaluation, and determining the effectiveness of drug matching by setting the drug adaptation threshold A, this method achieves precise matching between drugs and the patient's cause of illness. Traditional technologies often perform drug matching based on simple rules and empirical judgments, lacking systematic quantitative indicators. This method quantifies the recommendation validity of drugs by calculating the drug recommendation validity index. The biggest highlight of this technology is the introduction of gene compatibility analysis, which builds a bridge between drug recommendation and the patient's gene background. Through multi-dimensional analysis of gene drug efficacy data, the gene adaptation index jsp is obtained, enabling drug recommendation to not only be based on the cause of illness and indications but also take into account the patient's gene differences, further optimizing the personalized treatment plan. By comprehensively calculating the comprehensive gene adaptation index zht obtained from the drug recommendation validity index ytx and the gene adaptation index jsp, and performing a drug and gene matching evaluation with the preset gene adaptation threshold Z, the accuracy of drug recommendation and the treatment effect have been significantly improved. Especially in terms of the personalization of treatment effects and the safety of patient medication, this method has obvious advantages over traditional technologies, avoiding the occurrence of drug inadaptability, reducing risks during the treatment process, and ultimately improving the patient's treatment effect and medication experience. Compared with current traditional drug recommendation technologies, the biggest improvement of this method lies in comprehensively considering the patient's gene information and the multi-dimensional attributes of drugs, especially the comprehensive evaluation of drug ingredient correlation, indication matching degree, and gene adaptability.

[0071] Example 2

[0072] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1 , specifically: S1 includes S11, S12, and S13;

[0073] S11. The cloud platform collects drug ingredient data of each Western medicine from the Western medicine data source through the API dedicated interface. After cleaning and standardization, the natural language processing NLP technology is used to perform semantic parsing on the drug description;

[0074] The Western medicine data source includes a drug database, literature resources, clinical trials, drug regulatory agencies, and pharmaceutical factories;

[0075] S12. Use the cloud platform to build an intelligent data repository, structurally process the drug ingredient data, identify the drug ingredient attributes, including drug categories, drug ingredients, and drug efficacy, judge the sensitivity level of the drug according to the ingredient attributes, and then use the metadata index to label each drug based on the drug ingredient attributes and sensitivity level, and store the labeled drug ingredient data in the intelligent data repository;

[0076] The sensitivity levels are divided into anesthetic drugs, state-controlled drugs, and confidential drugs;

[0077] S13. The intelligent data repository enables server-side encryption, encrypts each drug data in the stored drug ingredient data with a corresponding unique encryption key, and uniformly manages it through the key management service KMS of the cloud platform.

[0078] In this embodiment, drug ingredient data is collected from multiple data sources through a dedicated API interface, and natural language processing technology is used to semantically analyze drug descriptions to ensure the accurate extraction and standardization of drug information. Secondly, after the drug ingredient data is structurally processed, relying on the intelligent data repository of the cloud platform, it can be managed in a labeled manner according to the ingredient attributes and sensitivity levels of the drugs, thereby improving the retrieval efficiency and operability of drug information. Finally, through server-side encryption and a unified key management service, the security and privacy protection of drug ingredient data are ensured, and the risk of sensitive data leakage is avoided. Overall, this module not only greatly improves the management efficiency and security of drug data, but also lays a solid foundation for subsequent drug self-recommendation platforms and gene adaptability analysis.

[0079] Embodiment 3

[0080] This embodiment is an explanatory description based on Embodiment 2. Please refer to Figure 1 , specifically: S2 includes S21 and S22;

[0081] S21. After the doctor diagnoses the patient's cause of illness and generates an electronic medical record, build a drug self-recommendation platform through the cloud platform. The drug self-recommendation platform uses medical natural language processing technology to segment and perform named entity recognition on the electronic medical record, map it to the standard cause of illness coding, establish a ternary map of diseases, symptoms, and drug ingredients, and locate all potential therapeutic ingredient drugs in the intelligent data repository corresponding to the cause of illness;

[0082] S22. The drug self-recommendation platform temporarily decrypts the located potential therapeutic ingredient drugs with a temporary key. The drug self-recommendation platform uses conditional combination retrieval based on the label attributes of each drug ingredient, extracts drugs suitable for specific pathological characteristics and individual conditions, recommends drugs for the patient, and extracts the information data of the drugs;

[0083] The information data includes drug characteristic dimensions and drug indications. The drug characteristic dimensions include pharmacodynamic mechanism, chemical structure, drug metabolism, tolerance, and dependence.

[0084] In this embodiment, through the construction of a drug self-recommendation platform and in-depth analysis of electronic medical records by combining medical natural language processing technology, this method significantly improves the accuracy and efficiency of the drug self-recommendation platform. Through word segmentation and named entity recognition, the cause information of the patient is mapped to the standard cause code, and a ternary graph of diseases, symptoms, and drug ingredients is established to achieve accurate matching of drugs and the patient's cause. Compared with traditional methods, it can not only automatically identify potential therapeutic ingredients but also perform conditional combination retrieval according to the label attributes and specific pathological characteristics of drugs to recommend the most suitable drugs for the patient's individual situation. This process fully considers the multi-dimensional characteristics of drugs, such as pharmacodynamic mechanism, chemical structure, drug metabolism, tolerance, and dependence, ensuring that the recommended drugs are more targeted and personalized, effectively reducing the mis-recommendation of unsuitable drugs and optimizing the patient's treatment plan. In addition, the use of a temporary key ensures data security and privacy protection, while enhancing flexibility and real-time performance, making drug recommendations not only accurate but also safe and reliable.

[0085] Example 4

[0086] This embodiment is an explanatory note for Embodiment 3. Please refer to Figure 1 and Figure 2 , specifically: S3 includes S31 and S32;

[0087] S31. After dimensionless processing of the drug ingredient correlation analysis and drug indication analysis based on the extracted drug information data, the drug ingredient correlation index ywg and the drug indication matching index spp are respectively calculated by summarization;

[0088] The drug ingredient correlation index ywg is obtained through the following formula and is used to analyze the correlation between drug ingredients;

[0089] ;

[0090] In the formula, ywg ij represents the correlation between drug i and drug j, n represents the number of characteristic dimensions, and w k represents the weight of the kth characteristic, and respectively represent the performance values of drug i and drug j in the k-th feature dimension, which are obtained from the drug instructions and normalized to values between 0 and 1. The specific values are set according to the actual situation. i and j represent index symbols, where i ∈ {1, 2,..., N} and j ∈ {1, 2,..., N}, indexing in the set of the total number of drugs N, and are used for one-to-one comparison between drugs;

[0091] The drug indication matching index spp is obtained through the following formula and is used to measure the matching degree of two drugs in terms of indications;

[0092] ;

[0093] In the formula, spp ij represents the drug indication matching index between drug i and drug j, represents the number of intersections of the indications of drug i and drug j, that is, the number of diseases jointly treated by drug i and drug j, represents the number of unions of the indications of drug i and drug j, that is, the number of all diseases treated by drug i and drug j.

[0094] S32. It is used to preliminarily judge the comprehensive suitability of the recommended drug and the patient's cause, specifically including S321 and S322;

[0095] S321. After dimensionless processing and summary calculation based on the obtained drug component correlation index ywg and drug indication matching index spp, the drug recommendation validity index ytx is obtained, which quantifies the comprehensive suitability of the recommended drug and the patient's cause;

[0096] ;

[0097] In the formula, cy i and cy j respectively represent the metabolic utility factors of drug i and drug j under specific patient conditions, represents the difference factor of the metabolic utility of drug i and drug j under specific patient conditions.

[0098] S322. Based on the drug recommendation validity indexes ytx of all drugs in the historical samples of the same type of cause, the mean u and standard deviation are calculated by the statistical method, and a preset drug suitability threshold A is set based on the mean u and standard deviation , specifically: , where, represents the confidence factor, which is used to adjust the sensitivity of the threshold, and then a preliminary drug matching evaluation is carried out with the obtained drug recommendation validity index ytx. The specific evaluation scheme is as follows;

[0099] When the drug recommendation validity index ytx > the drug adaptation threshold A, the drug matching is successful, and at this time, gene compatibility analysis is triggered;

[0100] When the drug recommendation validity index ytx ≤ the drug adaptation threshold A, the drug matching fails, and at this time, the drug self-recommendation platform automatically performs iterative optimization matching until the matching is successful.

[0101] In this embodiment, through the calculation and analysis of the drug component correlation index ywg and the drug indication matching index spp, this method can accurately quantify the correlation between drugs and their suitability for the patient's cause. The introduction of the drug recommendation validity index ytx not only provides a scientific basis for drug recommendation, but also quantifies the matching between drugs and the patient's specific cause, thus improving the accuracy of drug recommendation. On this basis, statistical analysis is carried out in combination with historical cause samples, and the drug adaptation threshold A is adjusted through the confidence factor, further improving the sensitivity and reliability of drug matching. This data-driven drug self-recommendation platform overcomes the deficiencies of traditional drug recommendation methods in personalized treatment, reduces the occurrence of drug inadaptation, and ensures the personalization and accuracy of the treatment plan. At the same time, the drug self-recommendation platform can automatically perform iterative optimization when the matching fails, greatly improving the intelligence and adaptability of the system, ensuring the best fit between the final drug recommendation and the patient's cause, and further optimizing the patient's treatment effect and medication experience.

[0102] Example 5

[0103] This embodiment is an explanatory description carried out in Example 4, please refer to Figure 1 , specifically: S4. When the preliminary drug matching evaluation is successful for drug matching, gene compatibility analysis is performed, which specifically includes S41 and S42;

[0104] S41. The gene compatibility analysis marks the successfully matched drug as drug X, and conducts a trial on the patient, and then obtains the gene pharmacodynamic data of the patient based on multi-dimensional data analysis technology;

[0105] The multi-dimensional data analysis technology includes whole-genome sequencing technology, electronic medical record system, and metabolic pathway-related gene expression analysis;

[0106] Using whole-genome sequencing technology to perform drug metabolism detection on the patient, matching the detection results with the metabolic pathway matching table in the intelligent data repository, and obtaining the gene adaptation factor sp by calculating the ratio of the matching channel to the total channel;

[0107] Extract the changes in physiological indicators after medication from the patient's electronic medical record system to obtain the physiological condition compatibility jr, and then compare it with the physiological indicators before medication to obtain the drug efficacy response yl;

[0108] Based on the gene expression analysis related to the patient's metabolic pathway, a drug metabolic pathway mapping diagram is established using an intelligent data repository. Through graph theory analysis of pathway node matching, the number of key enzymes with complete expression is compared with the total number of enzymes required for the drug to obtain the metabolic pathway matching value dx.

[0109] The gene drug effect data includes the gene adaptation factor sp, physiological condition compatibility jr, drug effect response yl, and metabolic pathway matching value dx.

[0110] S42. After dimensionless processing of the obtained gene drug effect data, summary calculations are performed to obtain the gene adaptability index jsp.

[0111] The gene adaptability index jsp is obtained through the following formula to analyze the compliance between the patient's medication needs and physiological status.

[0112] ;

[0113] In the formula, e represents the exponential function, m represents the total number of gene loci, p represents the gene locus number, represents the adaptation factor of drug X at the p-th gene locus, yl X represents the drug effect response of drug X to the patient, represents the metabolic pathway matching value of drug X on the p-th gene, jr X represents the physiological condition compatibility of drug X to the patient.

[0114] In this embodiment, based on multi-dimensional data analysis technologies such as whole-genome sequencing technology, electronic medical record system, and gene expression analysis related to metabolic pathways, the gene drug effect data of the patient can be accurately obtained, including the gene adaptation factor sp, physiological condition compatibility jr, drug effect response yl, and metabolic pathway matching value dx. After dimensionless processing of these data, the gene adaptability index jsp obtained by comprehensive calculation quantifies the adaptability between the drug and the patient's physiology and genes, thereby optimizing the accuracy and personalization level of drug matching. The special advantage of this method is that it not only considers the matching between the drug and the disease, but also takes into account the patient's genetic background, metabolic pathway, and physiological changes, making drug recommendations more accurate and having an efficient therapeutic effect. Finally, it effectively avoids the occurrence of drug inadaptation, improves the personalization of treatment plans, enhances the safety and efficacy of patient medication, and makes a positive contribution to the development of precision medicine.

[0115] Example 6

[0116] This embodiment is an explanatory description carried out in Example 5. Please refer to Figure 1 , specifically: S5 includes S51 and S52;

[0117] S51. Calculate comprehensively based on the drug recommendation validity index ytx and the gene suitability index jsp to obtain the comprehensive gene suitability index zht, which quantifies the suitability level of the drug required for gene matching.

[0118] The comprehensive gene suitability index zht is obtained through the following formula;

[0119] ;

[0120] In the formula, β represents the weight factor, and its value range is between 0 and 1. The specific value is set according to the actual situation and is used for the weight of the drug recommendation validity index ytx and the gene suitability index jsp in the comprehensive score.

[0121] S52. Preset the gene suitability threshold Z based on the gene suitability standard in the drug neighborhood, and conduct an evaluation on the drug-gene matching with the obtained comprehensive gene suitability index zht. The specific evaluation scheme is as follows;

[0122] When the comprehensive gene suitability index zht > the gene suitability threshold Z, it indicates that there is no hidden danger at the drug-gene level and it meets the patient's medication needs;

[0123] When the comprehensive gene suitability index zht ≤ the gene suitability threshold Z, it indicates that there is a hidden danger at the drug-gene level and it does not meet the patient's medication needs. At this time, the drug self-recommendation platform performs iterative optimization matching through S3 until the matching is successful.

[0124] In this embodiment, the comprehensive gene suitability index zht obtained by comprehensively calculating the drug recommendation validity index ytx and the gene suitability index jsp realizes the quantitative evaluation of the drug-gene level matching. The introduction of the comprehensive gene suitability index makes drug recommendation not only limited to the matching of indications and causes, but also emphasizes the personalized differences in the patient's gene background, significantly improving the accuracy and safety of drug recommendation. By setting the gene suitability threshold Z, it can effectively distinguish the suitability of drugs and patients' genes, ensuring that only drugs that meet the patient's gene characteristics can enter the recommendation list, thus avoiding the potential risk of gene-drug mismatch. When the comprehensive gene suitability index zht is higher than the preset threshold, it is considered that there is no hidden danger at the drug-gene level and it meets the patient's medication needs, ensuring the personalization and effectiveness of the treatment plan. When zht is lower than this threshold, iterative optimization is automatically triggered until a drug that meets the patient's gene needs is found. This process of dynamic optimization and precise matching greatly improves the applicability and treatment effect of the drug self-recommendation platform in practical applications, ensures the medication safety of patients, and at the same time improves the level of personalized treatment.

[0125] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent storage and management method for western medicine pharmaceutical ingredient data, characterized in that: It includes the following steps: S1. Build an intelligent data repository based on the cloud platform, and then collect the drug ingredient data of Western medicines from the Western medicine data source through the API dedicated interface. After semantic analysis and tagging, it is encrypted and stored in the intelligent data repository; S2. The cloud platform builds a drug self-recommendation platform, locates the drugs of all potential therapeutic ingredients in the intelligent data repository through the electronic medical records of patients, decrypts them temporarily, retrieves them according to the usage conditions of label attributes in combination, and extracts the information data of the drugs; S3. Conduct drug ingredient correlation analysis and drug indication analysis based on the extracted drug information data, and perform summary calculations based on the analysis results to obtain the drug recommendation validity index ytx for preliminary drug matching evaluation. If the drug recommendation validity index ytx exceeds the drug adaptation threshold A, then perform gene compatibility analysis; S4. After receiving the gene compatibility analysis instruction, obtain the gene drug efficacy data of the patient based on the multi-dimensional data analysis technology, and based on the gene drug efficacy data, analyze the compliance of the patient's medication needs and physiological state to obtain the gene adaptability index jsp; S5. Perform comprehensive calculations on the drug recommendation validity index ytx and the gene adaptability index jsp to obtain the comprehensive gene adaptability index zht for medication and gene matching evaluation.

2. The intelligent storage management method for western medicine pharmaceutical ingredient data according to claim 1, characterized in that: S1 includes S11, S12 and S13; S11. The cloud platform collects the drug ingredient data of each Western medicine from the Western medicine data source through the API dedicated interface. After cleaning and standardization, the natural language processing NLP technology is used to perform semantic analysis on the drug description; The Western medicine data source includes drug databases, literature resources, clinical trials, drug regulatory agencies and pharmaceutical factories; S12. Use the cloud platform to build an intelligent data repository, structurally process the drug ingredient data, identify the drug ingredient attributes, including drug categories, drug ingredients and drug efficacy, and judge the sensitivity level of the drug according to the ingredient attributes. Then, use the metadata index to tag each drug according to the drug ingredient attributes and sensitivity level, and store the tagged drug ingredient data in the intelligent data repository; The sensitivity levels are divided into anesthetic drugs, state-controlled drugs and confidential drugs; S13. The intelligent data repository enables server-side encryption, encrypts each drug data in the stored drug ingredient data with a corresponding unique encryption key, and is uniformly managed through the key management service KMS of the cloud platform.

3. The intelligent storage management method for western medicine pharmaceutical ingredient data according to claim 2, characterized in that: S2 includes S21 and S22; S21. After the doctor diagnoses the patient's cause of illness and generates an electronic medical record, build a drug self-recommendation platform through the cloud platform. The drug self-recommendation platform uses medical natural language processing technology to segment the electronic medical record and identify named entities, map them to the standard cause of illness coding, establish a ternary graph of disease, symptom and drug ingredient, and locate all potential therapeutic ingredient drugs in the intelligent data repository corresponding to the cause of illness; S22. The drug self-recommendation platform temporarily decrypts the drugs of the potentially therapeutic ingredients located through the temporary key. The drug self-recommendation platform retrieves according to the combination of usage conditions with label attributes for each drug ingredient, extracts the drugs suitable for specific pathological features and individual conditions, recommends drugs for the patient, and extracts the information data of the drugs. The information data includes the drug feature dimension and the drug indication.

4. The intelligent storage management method for western medicine pharmaceutical ingredient data according to claim 3, characterized in that: S3 includes S31 and S32; S31. After the dimensionless processing of the drug ingredient correlation analysis and the drug indication analysis based on the extracted drug information data, the drug ingredient correlation index ywg and the drug indication matching index spp are respectively obtained by summarizing and calculating. The drug ingredient correlation index ywg is obtained through the following formula; ； where, ywg ij represents the association degree between drug i and drug j, n represents the number of feature dimensions, and w k represents the weight of the k-th feature, and respectively represent the performance values of drug i and drug j on the k-th feature dimension. i and j are index symbols, where i ∈ {1, 2,..., N} and j ∈ {1, 2,..., N}, indexing within the set of the total number of drugs N extracted; The drug indication matching index spp is obtained through the following formula; ； In the formula, spp ij represents the drug indication matching index between drug i and drug j, represents the number of intersections of the indications of drug i and drug j, represents the number of unions of the indications of drug i and drug j.

5. A method for intelligent storage management of western medicine pharmaceutical ingredient data according to claim 4, characterized in that: S32. It is used to preliminarily judge the comprehensive suitability of the recommended drug and the patient's cause, specifically including S321 and S322; S321. After the dimensionless processing of the obtained drug ingredient correlation index ywg and the drug indication matching index spp, they are summarized and calculated to obtain the drug recommendation validity index ytx, which quantifies the comprehensive suitability of the recommended drug and the patient's cause. ； wherein, cy i and cy j respectively represent the metabolic utility factors of drug i and drug j under specific patient conditions, represents the difference factor of the metabolic utility of drug i and drug j under specific patient conditions.

6. The intelligent storage and management method for western medicine pharmaceutical ingredient data according to claim 5, wherein: S322. All drug recommendation validity indices ytx based on historical samples of the same type of etiology are used to calculate the mean value u and standard deviation through statistical methods , and based on the mean value u and standard deviation , a preset drug adaptation threshold A is set, specifically as follows: , where represents the confidence factor, which is used to adjust the sensitivity of the threshold, and then a preliminary drug matching evaluation is carried out with the obtained drug recommendation validity index ytx. The specific evaluation plan is as follows; When the drug recommendation validity index ytx > the drug adaptation threshold A, the drug matching is successful, and at this time, the gene compatibility analysis is triggered; When the drug recommendation validity index ytx ≤ the drug adaptation threshold A, the drug matching fails, and at this time, the drug self-recommendation platform automatically performs iterative optimization matching until the matching is successful.

7. The intelligent storage management method for western medicine pharmaceutical ingredient data according to claim 6, characterized in that: S4. It is used to perform gene compatibility analysis when the preliminary drug matching evaluation is successful for drug matching, specifically including S41 and S42; S41. The gene compatibility analysis marks the successfully matched drug as drug X, and conducts a trial on the patient, and then obtains the gene drug efficacy data of the patient according to the multi-dimensional data analysis technology. The multi-dimensional data analysis technology includes whole genome sequencing technology, electronic medical record system, and metabolic pathway-related gene expression analysis; Using the whole genome sequencing technology to perform drug metabolism detection on the patient, matching the detection results with the metabolic pathway matching table in the intelligent data repository, and obtaining the gene adaptation factor sp by calculating the ratio of the matching channel to the total channel; Extract the changes in physiological indicators after medication from the patient's electronic medical record system to obtain the physiological condition compatibility jr, and then compare it with the physiological indicators before medication to obtain the drug efficacy response yl; Based on the metabolic pathway-related gene expression analysis of the patient, use the intelligent data repository to establish a drug metabolic pathway mapping diagram, analyze the matching of pathway nodes through graph theory, and compare the number of key enzymes with complete expression with the total number of enzymes required for the drug to obtain the metabolic pathway matching value dx; The gene drug efficacy data includes the gene adaptation factor sp, the physiological condition compatibility jr, the drug efficacy response yl, and the metabolic pathway matching value dx.

8. A method for intelligent storage management of western medicine pharmaceutical ingredient data according to claim 7, characterized in that: S42. After the dimensionless processing of the obtained gene drug efficacy data, they are summarized and calculated to obtain the gene adaptability index jsp; The gene adaptability index jsp is obtained through the following formula to analyze the compliance of the patient's medication needs and physiological status; ； In the formula, e represents the exponential function, m represents the total number of gene loci, p represents the gene locus number, represents the adaptation factor of drug X at the p-th gene locus, yl X represents the pharmacodynamic response of drug X to the patient, represents the metabolic pathway matching value of drug X on the p-th gene, jr X represents the physiological condition compatibility of drug X with the patient.

9. The intelligent storage and management method for western medicine pharmaceutical ingredient data according to claim 1, characterized in that: S5 includes S51 and S52; S51. Based on the comprehensive calculation of the drug recommendation validity index ytx and the gene adaptability index jsp, the comprehensive gene adaptability index zht is obtained, which quantifies the adaptability level of the drug required for gene matching; The comprehensive gene adaptability index zht is obtained through the following formula; ； In the formula, β represents the weight factor.

10. The intelligent storage and management method for western medicine pharmaceutical ingredient data according to claim 9, characterized in that: S52. Based on the preset gene adaptability threshold Z for the drug neighborhood gene adaptability standard, the drug-gene matching evaluation is carried out with the obtained comprehensive gene adaptability index zht. The specific evaluation scheme is as follows; When the comprehensive gene adaptability index zht > the gene adaptability threshold Z, it indicates that there is no hidden danger at the drug-gene level and it meets the patient's medication needs; When the comprehensive gene adaptability index zht ≤ the gene adaptability threshold Z, it indicates that there is a hidden danger at the drug-gene level and it does not meet the patient's medication needs. At this time, the drug is iteratively optimized and matched through S3 on the self-recommendation platform until the matching is successful.

Images