Test data management method and system for traditional Chinese medicine innovation

Abstract

The application discloses a test data management method and system for traditional Chinese medicine innovation, and relates to the field of data management. The method comprises the following steps: receiving, through a user terminal, research and development demand information of a traditional Chinese medicine innovation project, wherein the research and development demand information comprises a target disease and an efficacy index; connecting a data sharing platform, searching for multi-party test data based on the target disease, and obtaining a multi-party test data set; performing medicinal material correlation analysis on the multi-party test data set, determining a candidate medicinal material list matched with the target disease, and determining a candidate medicinal material proportion interval set; performing constraint optimization on the candidate medicinal material proportion interval set based on the efficacy index, and determining an optimal medicinal material proportion interval set meeting the efficacy index; and generating a test data management report of the traditional Chinese medicine innovation project according to the candidate medicinal material list and the optimal medicinal material proportion interval set, and sending the test data management report to the user terminal. The problems of low success probability of traditional Chinese medicine research and development, inaccurate medicinal material evaluation results and poor efficacy are solved.

Description

Technical Field

[0001] This invention relates to the field of data management, and more specifically to a method and system for managing test data for innovation in traditional Chinese medicine. Background Technology

[0002] In existing technologies, the innovation testing of traditional Chinese medicine relies heavily on human experience for judgment, which makes the selection criteria for medicinal materials susceptible to subjective factors. This leads to uncertainty in the efficacy of medicinal materials, resulting in a low success rate in the research and development of traditional Chinese medicine materials, as well as problems such as incorrect use of medicinal materials and poor efficacy.

[0003] In conclusion, there is a need for a scientific method to evaluate innovative TCM testing data, utilizing multi-source data to quantitatively evaluate medicinal materials, thereby improving the success rate of research and development and the accuracy of evaluation results. Summary of the Invention

[0004] This application provides a test data management method and system for innovation in traditional Chinese medicine, aiming to solve the problems in the existing technology that rely on human experience for judgment, which makes the selection criteria of medicinal materials susceptible to subjective factors, and the efficacy of medicinal materials uncertain, resulting in a low probability of successful research and development of traditional Chinese medicine materials, as well as the problems of incorrect medicinal materials and poor efficacy.

[0005] In view of the above problems, this application provides a test data management method and system for innovation of traditional Chinese medicine.

[0006] Firstly, this application provides a method for managing test data for innovation in traditional Chinese medicine, including: The system receives research and development demand information for innovative traditional Chinese medicine projects through a user terminal. The research and development demand information includes target diseases and efficacy indicators. Connect to the data sharing platform, and retrieve multi-party test data based on the target disease to obtain a multi-party test dataset; Perform medicinal material correlation analysis on the multi-party test dataset to determine a list of candidate medicinal materials that match the target disease, and determine the set of candidate medicinal material ratio intervals; Based on the efficacy indicators, the set of candidate medicinal material ratio intervals is constrained and optimized to determine the set of preferred medicinal material ratio intervals that meet the efficacy indicators. A test data management report for the innovative traditional Chinese medicine project is generated based on the candidate medicinal materials list and the preferred medicinal material ratio range set, and then sent to the user terminal.

[0007] Secondly, this application provides a test data management system for innovation in traditional Chinese medicine, including: The demand information receiving module is used to receive research and development demand information of traditional Chinese medicine innovation projects through the user terminal. The research and development demand information includes target diseases and efficacy indicators. The multi-party test data retrieval module is used to connect to the data sharing platform, perform multi-party test data retrieval based on the target disease, and obtain a multi-party test dataset; The correlation analysis module is used to perform medicinal material correlation analysis on the multi-party test dataset, determine the list of candidate medicinal materials that match the target disease, and determine the set of candidate medicinal material ratio intervals; The ratio range optimization module is used to perform constraint optimization on the candidate medicinal material ratio range set based on the efficacy index, and determine the preferred medicinal material ratio range set that meets the efficacy index. The test report generation module is used to generate a test data management report for the traditional Chinese medicine innovation project based on the candidate medicinal material list and the preferred medicinal material ratio range set, and send it to the user terminal.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application transforms R&D needs into concrete solutions by establishing a complete data processing chain. It utilizes a data sharing platform for data retrieval, effectively integrating multi-source data and resolving the issue of incomplete data. Through quantitative evaluation of the compatibility and efficacy of medicinal materials, objective screening standards for medicinal materials are established, avoiding biases caused by subjective experience. A reverse optimization mechanism proactively identifies and avoids ineffective combinations, increasing the probability of successful R&D. This achieves a data-driven transformation in the evaluation of traditional Chinese medicine, providing scientific and reliable technical support for innovative R&D in this field. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A flowchart illustrating a test data management method for innovation in traditional Chinese medicine; Figure 2 This is a schematic diagram of the structure of a test data management system for innovation in traditional Chinese medicine.

[0011] In the attached diagram, the labels represent the following: 11 Requirement Information Receiving Module; 12 Multi-Party Test Data Retrieval Module; 13 Correlation Analysis Module; 14 Proportioning Range Optimization Module; 15 Test Report Generation Module. Detailed Implementation

[0012] This application provides a test data management method for innovation in traditional Chinese medicine, which addresses the problem that existing technologies rely on human experience for judgment, making the selection criteria for medicinal materials susceptible to subjective factors, resulting in uncertain efficacy of medicinal materials, low success rate in the research and development of traditional Chinese medicine materials, and poor efficacy.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0015] like Figure 1 As shown, this application provides a method for managing test data for innovation in traditional Chinese medicine, the method comprising: S10: Receive research and development demand information for traditional Chinese medicine innovation projects through the user terminal, wherein the research and development demand information includes target diseases and efficacy indicators; In this embodiment, the user terminal is a web interface or a mobile application. The target disease refers to the specific disease requiring treatment. The efficacy indicator refers to the quantitative standard for the expected treatment effect.

[0016] By collecting target diseases and efficacy indicators input by users through the user terminal, the information is recorded and parsed, and standardized into a queryable format as the basis for subsequent data retrieval and analysis.

[0017] For example, if a plan is to develop a traditional Chinese medicine for lowering blood sugar, the user inputs the target disease as type 2 diabetes and the efficacy indicator as a 1.5% reduction in glycated hemoglobin (HbA1c). The system parses the information and standardizes it into ICD-10 disease codes and efficacy units.

[0018] In this embodiment of the application, by specifying the R&D requirements into operable technical parameters, clear input conditions are provided for subsequent data processing, thereby realizing the quantitative management of R&D objectives.

[0019] To address the above issues, this application connects to a data sharing platform to retrieve multi-party test data based on the target disease, thereby obtaining a multi-party test dataset.

[0020] S20: Connect to the data sharing platform, perform multi-party test data retrieval based on the target disease, and obtain a multi-party test dataset; In this embodiment of the application, the data sharing platform refers to a unified access interface that integrates multiple data sources, and the multi-party test data refers to scientific research and clinical data from different sources.

[0021] By connecting to data sharing platforms and integrating scattered data, and connecting to these platforms via APIs or database interfaces, relevant data can be retrieved based on the target disease. This establishes a unified data access channel, breaks down data silos, and achieves effective integration of multi-source data.

[0022] To address the above issues, this application connects to a data sharing platform to retrieve multi-party test data based on the target disease, thereby obtaining a multi-party test dataset.

[0023] Step S20 in the method provided in this application embodiment includes: Connect to a data sharing platform, which includes multiple data storage terminals; Data retrieval conditions are constructed based on the target disease, and data retrieval requests are sent to the multiple data storage terminals based on the data retrieval conditions; Multiple data storage terminals respectively return multiple de-identified datasets based on the data retrieval request; The multiple de-identified datasets are aggregated to obtain the multi-party test dataset.

[0024] In this embodiment, the platform is first connected to multiple data storage endpoints via a standard connection protocol. Each endpoint may use a different database type. These multiple data storage endpoints represent a platform composed of multiple independent data sources, such as databases from hospitals, research institutions, and pharmaceutical companies.

[0025] Secondly, based on the characteristics of the target disease, a search expression is generated to construct data retrieval conditions. Then, a data retrieval request is sent to the data storage terminal based on these conditions to ensure query accuracy. Specifically, the construction of search conditions involves generating a structured query based on the target disease.

[0026] Furthermore, each of the multiple data storage terminals executes a query locally based on the data retrieval request, and the results are anonymized.

[0027] Finally, the anonymized datasets from each data source are deduplicated and standardized, and then the multiple anonymized datasets are aggregated and stored in a structured format to obtain a multi-source test dataset. Data aggregation and merging refers to the formation of a unified dataset from multiple data sources.

[0028] For example, 2,000 prescription records are obtained from three data sources. After summarizing and removing duplicate records, 1,500 data records are finally obtained.

[0029] In this embodiment of the application, the data retrieval conditions for the target disease include: Extract disease feature information of the target disease, including disease classification, pathogenesis, and clinical symptoms; Based on the disease classification, pathogenesis, and clinical symptoms, a search expression is constructed as the data retrieval condition.

[0030] In this embodiment of the application, the disease features are first extracted by analyzing the multi-dimensional features of the disease based on the medical knowledge base. The medical knowledge base can be called to analyze the classification, mechanism and symptoms of the target disease.

[0031] For example, the target disease is diabetes, and the feature extraction results are: the disease is classified as a chronic disease, the pathogenesis is hyperglycemia caused by peripheral insulin resistance, and the clinical symptoms are polydipsia, polyphagia, and blurred vision.

[0032] Secondly, based on disease classification, pathogenesis, and clinical symptoms, search expressions are constructed as data retrieval conditions. The search expression generation process transforms features into a query language that the database can understand.

[0033] For example, features are converted into logical expressions, then into SQL, and keywords are connected using Boolean operators (AND / OR).

[0034] In this embodiment of the application, the data retrieval request returns multiple de-identified datasets, including: The first data storage terminal is determined from the plurality of data storage terminals; The first data storage terminal retrieves matching data from the local database according to the data retrieval request to obtain the first matching dataset. The first matching dataset is anonymized to generate a first anonymized dataset, which is then returned. Following the method used to obtain the first de-identified dataset, the other data storage terminals synchronously obtain the corresponding de-identified datasets, resulting in multiple de-identified datasets.

[0035] In this embodiment, the first data storage terminal is determined from a data storage terminal according to the configuration order such as data source priority, and then the request that is at the beginning of the configuration order is sent first.

[0036] For example, the hospital database is connected first, sorted by priority.

[0037] Secondly, after receiving a query request, the first data storage terminal retrieves matching data from its local database. The retrieved data serves as the first matching dataset. Notably, the local retrieval is performed independently by each data storage terminal, without exposing the original data.

[0038] Next, an anonymization algorithm was applied to the results of the first matching dataset to generate the first anonymized dataset. Multiple anonymized datasets contained research data such as medicinal materials, formulations, and therapeutic effects. The anonymization process removed personal privacy information, retaining only research-related data, including replacing patient information with encoded data.

[0039] For example, the hospital database retrieves 100 records. After desensitization, the original data returned contains the patient's age of 45 years old. After desensitization, the data becomes the age group of 40-50 years old and the anonymous ID is P001.

[0040] Desensitization rules are based on predefined rules for processing sensitive information.

[0041] Finally, following the same method used to obtain the first de-identified dataset, parallel or serial requests are made to the remaining data storage terminals to synchronously obtain the corresponding de-identified datasets, resulting in multiple de-identified datasets.

[0042] For example, anonymized data is retrieved sequentially from hospital databases and pharmaceutical company databases, and finally merged into a dataset containing 500 records.

[0043] In this embodiment, firstly, multiple data storage terminals are connected to build a data sharing platform, providing multi-source data. Secondly, disease characteristic information of the target disease is extracted, a retrieval expression is constructed, a unified expression pattern is formed, and intelligent distribution is performed to ensure data processability. Subsequently, data retrieval requests are sent to multiple data storage terminals according to the data retrieval conditions to improve the security of privacy data.

[0044] In addition, a primary data storage terminal is determined from multiple data storage terminals, and matching data is retrieved from the local database to obtain the first matching dataset. Through anonymization processing, the anonymized dataset is obtained synchronously, ensuring efficient data acquisition while maintaining data availability and privacy. Finally, the multiple anonymized datasets are integrated to obtain a multi-party test dataset, forming correlated structured data that is beneficial for statistical analysis.

[0045] S30: Perform medicinal material correlation analysis on the multi-party test dataset to determine a list of candidate medicinal materials that match the target disease, and determine the set of candidate medicinal material ratio intervals; In this embodiment, the applicable correlation analysis refers to the degree of matching between medicinal materials and diseases. The candidate list is generated by screening high-value medicinal materials using a correlation threshold.

[0046] By using correlation analysis, a list of medicinal materials is extracted from the dataset and the correlation degree is calculated to ensure the scientific validity of the candidate medicinal materials and obtain initial evaluation results.

[0047] To address the above issues, this application performs a medicinal material correlation analysis on the multi-party test dataset to determine a list of candidate medicinal materials that match the target disease, and to determine a set of candidate medicinal material ratio intervals.

[0048] Step S30 in the method provided in this application embodiment includes: Extract a complete list of medicinal materials from the multi-party test dataset. The complete list of medicinal materials includes multiple Chinese medicines and the range of medicinal material ratios for each Chinese medicine. An applicability correlation analysis is performed on the multiple traditional Chinese medicines and the target disease to obtain multiple applicability correlations. Based on the multiple applicability correlations, the multiple traditional Chinese medicines are screened to obtain a candidate medicinal material list, which includes multiple candidate traditional Chinese medicines. Extract the ratio ranges of multiple candidate Chinese medicinal materials and construct a set of candidate medicinal material ratio ranges corresponding to the candidate medicinal material list.

[0049] In this embodiment, firstly, the entire multi-party test dataset is traversed to statistically analyze the frequency of occurrence and the ratio range of each medicinal material. The complete list of medicinal materials consists of all the medicinal materials in the multi-party test database and their ratio ranges. The ratio range is the difference between the maximum and minimum dosage of the medicinal material.

[0050] For example, from 1500 prescription data, the list of medicinal materials is extracted as follows: Astragalus membranaceus: appeared 800 times, with a ratio range of 5-40 grams; Rehmannia glutinosa: appeared 825 times, with a ratio range of 5-30 grams; Coptis chinensis: appeared 1080 times, with a ratio range of 3-20 grams.

[0051] Secondly, an applicability correlation analysis was conducted on multiple traditional Chinese medicines (TCMs) in the list and the target diseases, resulting in multiple applicability correlations. Taking into account historical usage frequency and efficacy data, multiple TCMs were screened based on these applicability correlations to obtain a candidate medicinal materials list. This candidate medicinal materials list includes multiple candidate TCMs.

[0052] Finally, candidate medicinal materials were screened using correlation thresholds, and their ratio ranges were collected. The ratio ranges of multiple candidate Chinese medicinal materials were extracted to construct a set of candidate medicinal material ratio ranges corresponding to the candidate medicinal material list.

[0053] For example, the candidate medicinal materials list is: Coptis chinensis, Astragalus membranaceus, and Rehmannia glutinosa. Based on the corresponding medicinal material ratio ranges, a set of candidate medicinal material ratio ranges is constructed: [Coptis chinensis, ratio range: 3-20g; Rehmannia glutinosa, ratio range: 5-30g; Astragalus membranaceus, ratio range: 5-40g].

[0054] In this embodiment of the application, the list of candidate medicinal materials includes multiple candidate traditional Chinese medicines, including: Extract multiple historical finished drugs from the multi-party test dataset to obtain the total number of finished drugs; The frequency of use of each of the aforementioned traditional Chinese medicines as a preparatory ingredient in the multiple historical finished medicines is statistically analyzed, and the ratio of the frequency of use to the total number of finished medicines is calculated to obtain the compatibility support of each of the aforementioned traditional Chinese medicines. For each of the aforementioned traditional Chinese medicines, the efficacy satisfaction of historical finished products containing the aforementioned traditional Chinese medicines is extracted, and the average of the efficacy satisfaction is calculated to obtain the efficacy support level of each of the aforementioned traditional Chinese medicines. Based on the compatibility support and efficacy support of each of the Chinese herbal medicines, the applicability correlation between each of the Chinese herbal medicines and the target disease is calculated; A preset correlation threshold is set, and Chinese medicines with a correlation greater than the preset correlation threshold are screened to obtain multiple candidate Chinese medicines and generate a candidate medicinal material list.

[0055] In this embodiment, firstly, multiple historical finished drugs are extracted from a multi-party test dataset to obtain the total number of finished drugs. Secondly, the frequency of use of each traditional Chinese medicine (TCM) as a preparatory drug in the multiple historical finished drugs is statistically analyzed, and the ratio of the frequency of use to the total number of finished drugs is calculated to obtain the compatibility support of each TCM. Compatibility support = Frequency of use / Total number of finished drugs. The compatibility degree can initially reflect the applicability of TCM as a preparatory drug to the target disease.

[0056] For example, according to statistics, Astragalus was used 800 times in type 2 diabetes, and there were a total of 1,500 finished drugs. The compatibility support of Astragalus = 800 / 1,500 = 0.53.

[0057] Furthermore, for each traditional Chinese medicine (TCM), the efficacy satisfaction rates of historical finished products containing the TCM were extracted from the database. The average of these satisfaction rates was calculated to obtain the efficacy support rate for each TCM. The calculation formula is: Efficacy Support Rate = Sum of Efficacy Satisfaction Rates / Number of Efficacy Satisfaction Rates. The efficacy support rate reflects the curative effect of historical finished products containing the TCM on the target disease; the higher the value, the better the curative effect.

[0058] For example, in type 2 diabetes, the satisfaction rate of Astragalus membranaceus was 0.85, 0.8, and 0.9. The efficacy support rate was (0.85 + 0.8 + 0.9) / 3 = 0.85, which means the efficacy support rate was 85%.

[0059] Furthermore, based on the compatibility support and efficacy support of each traditional Chinese medicine (TCM), the applicability correlation between each TCM and the target disease is calculated. The applicability correlation is a weighted fusion of historical usage frequency and efficacy data, and the calculation formula is: Usage Correlation = Weight × Historical Usage Frequency + Weight × Efficacy Satisfaction. The weights can be set according to the degree of influence of historical usage frequency and efficacy satisfaction on the usage correlation.

[0060] For example, the weights for historical usage frequency and therapeutic satisfaction are set to 0.6 and 0.4, respectively. Correlation = 0.6 × historical usage frequency + 0.4 × therapeutic satisfaction = 0.6 × 0.53 + 0.4 × 0.85 = 0.658.

[0061] Finally, a preset correlation threshold is set, and Chinese medicinal herbs with a correlation greater than the preset threshold are selected to obtain multiple candidate herbs and generate a candidate herb list. The correlation threshold should not be set too high; a range of 0.5 to 0.8 is recommended. Users can configure this threshold according to their actual business scenarios. If no user configures it, the default value is 0.5.

[0062] For example, if the correlation threshold is 0.5, and the correlation of Astragalus membranaceus is 0.658 > 0.5, then Astragalus membranaceus will be included in the candidate medicinal material list. Meanwhile, the correlations of Coptis chinensis and Rehmannia glutinosa are 0.72 and 0.55 respectively, which meet the threshold. Therefore, the candidate medicinal material list is obtained as follows: Coptis chinensis, Rehmannia glutinosa, and Astragalus membranaceus.

[0063] In this embodiment, a complete list of medicinal materials is first extracted from a multi-party test dataset to construct a comprehensive data list, avoiding the omission of potentially valuable medicinal materials. Compatibility support and efficacy support are obtained from the multi-party test dataset, and applicability correlation analysis is performed. The quantified applicability correlation improves the comparability of medicinal material evaluations. Secondly, based on the applicability correlation calculation and a preset correlation threshold, a list of medicinal materials strongly correlated with and applicable to the target disease is obtained. Finally, a set of candidate medicinal material ratio intervals corresponding to the candidate medicinal material list is constructed to ensure quantitative output results and greatly reduce the blindness of testing.

[0064] S40: Based on the efficacy indicators, perform constraint optimization on the candidate medicinal material ratio interval set to determine the preferred medicinal material ratio interval set that meets the efficacy indicators; In this embodiment, the candidate medicinal material ratio range is constrained and optimized based on efficacy indicators. Through inverse optimization, high-risk medicinal material ratio ranges are eliminated, and the optimal medicinal material ratio range that meets the efficacy indicators is determined. This improves the scientific nature of the decision-making process.

[0065] To address the above issues, this application optimizes the candidate medicinal material ratio range based on the efficacy indicators to determine the preferred medicinal material ratio range that meets the efficacy indicators.

[0066] Step S40 in the method provided in this application embodiment includes: A threshold for treatment satisfaction is set based on the aforementioned treatment indicators; From the historical finished drug data set of the multi-party test dataset, finished drugs with efficacy satisfaction lower than the efficacy satisfaction threshold are selected and identified as the inefficient finished drug set. Extract the proportion data of each candidate Chinese medicine contained in the inefficient finished drug set and establish a database of poor proportions; The candidate medicinal material ratio range set is compared with the undesirable ratio database, and overlapping ratio ranges are eliminated; The eliminated ratio ranges are integrated to generate a set of optimal medicinal material ratio ranges.

[0067] In this embodiment, the efficacy satisfaction index is a numerical standard that can be used to screen historical data, and can be obtained through data such as clinical effectiveness rate, symptom improvement rate, and biochemical indicator compliance rate. The efficacy satisfaction threshold can be set and can directly reflect the expected value of drug efficacy.

[0068] When obtaining the optimal range of medicinal herb ratios, a threshold for efficacy satisfaction is first set based on efficacy indicators. This threshold is then used as the numerical standard for filtering historical data. A higher threshold results in a more conservative optimized ratio range, but also a higher expected success rate; a lower threshold results in a wider range of retained ratios, but also increases uncertainty. Therefore, the efficacy satisfaction threshold should not be too large, and a range of 0.6 to 0.8 is recommended. Users can configure this threshold according to their actual business scenarios. If no configuration is made, the default value is 0.6.

[0069] For example, the input efficacy indicator is: a new traditional Chinese medicine for treating type 2 diabetes, with a reduction of glycated hemoglobin (HbA1c) of more than 1.0%.

[0070] Secondly, from the historical finished drug data sets of multiple testing datasets, finished drugs with efficacy satisfaction rates below the efficacy satisfaction threshold were screened to identify the inefficient finished drug set. This screening process centralizes scattered failure experiences, laying a data foundation for subsequent root cause analysis.

[0071] For example, the efficacy satisfaction threshold is 0.6. 1500 ineffective finished drugs with efficacy satisfaction scores less than 0.6 are obtained, forming a set of ineffective finished drugs.

[0072] Furthermore, the formulation data of each candidate Chinese herbal medicine in the ineffective finished drug set are extracted, and the formulation data of all ineffective finished drugs are analyzed to establish an undesirable formulation database. This database records the dosage and proportion of candidate medicinal materials used in specific amounts, resulting in poor efficacy.

[0073] For example, an analysis of 1500 ineffective finished medicines was conducted to extract the proportions of candidate medicinal materials. It was found that in ineffective prescriptions, the dosage of Astragalus membranaceus (Huang Qi) was concentrated in two ranges: below 10 grams and above 40 grams. When the dosage of Coptis chinensis (Huang Lian) exceeded 15 grams, there were frequent records of poor efficacy accompanied by gastrointestinal discomfort. Concomitant use generally resulted in poor efficacy.

[0074] The database of undesirable combinations was constructed as follows: {Medicinal herb: Astragalus membranaceus, undesirable range: [5, 10]}, {Medicinal herb: Astragalus membranaceus, undesirable range: [40, 60]}, {Medicinal herb: Coptis chinensis, undesirable range: [15, 20]}, {Medicinal herb combination: (Astragalus membranaceus, Coptis chinensis), undesirable condition: dosage of Astragalus membranaceus > dosage of Coptis chinensis}.

[0075] Furthermore, the candidate medicinal material ratio range set was compared with the database of poor ratios. Ranges that overlapped with the database of poor ratios were eliminated due to high failure risk, which improved the targeting and success rate of subsequent tests.

[0076] Finally, by comparing with the database of undesirable ratios, the ratio ranges of medicinal materials that have been initially screened and eliminated are obtained. The ratio ranges after elimination are integrated to generate a set of optimal medicinal material ratio ranges.

[0077] In this embodiment, the process begins by screening finished drugs whose efficacy satisfaction is below a threshold. Next, formulation data is extracted to establish a database of undesirable formulations. Reverse screening is then used for efficient risk optimization. The candidate formulation range set is then compared with the database of undesirable formulations to eliminate overlapping ranges, avoid known risks, reduce unsafe ranges, and improve the targeting and success rate of subsequent tests. Through a closed-loop testing process involving threshold setting, identification, feature extraction, comparison and elimination, and integrated output, a reliable optimized formulation range set is obtained, improving R&D success rate and reducing risk and cost.

[0078] S50: Generate a test data management report for the traditional Chinese medicine innovation project based on the candidate medicinal materials list and the preferred medicinal material ratio range set, and send it to the user terminal.

[0079] In this embodiment, based on the candidate medicinal material list and the preferred medicinal material ratio range, all analysis results are integrated to generate a test data management report for the traditional Chinese medicine innovation project, which is then sent to the user terminal, completing the data analysis loop. The test data management report includes the data analysis process, conclusions, and risk warnings.

[0080] Test data management reports can provide specific optimization suggestions and solutions to improve the effectiveness of subsequent research and testing.

[0081] The embodiments of this application, through the specific implementation methods described above, achieve the following technical effects: In this embodiment, the R&D requirements are first concretized into operable technical parameters, providing clear input conditions for subsequent data processing, thereby achieving quantitative management of R&D goals. Secondly, multiple data storage terminals are connected to build a data sharing platform, providing multi-source data. Disease characteristic information of the target disease is extracted, and based on disease classification, pathogenesis, and clinical symptoms, a retrieval expression is constructed as a data retrieval condition. Data retrieval requests are then sent to multiple data storage terminals based on these conditions. By constructing the retrieval expression, a unified expression pattern is formed for intelligent distribution, ensuring data processability.

[0082] Furthermore, a primary data storage terminal is determined from multiple data storage terminals, and matching data is retrieved from the local database to obtain the first matching dataset. Through data anonymization, multiple anonymized datasets are obtained, efficiently acquiring data while ensuring its availability and privacy. Integrating these anonymized datasets yields a multi-party test dataset, which facilitates statistical analysis and the formation of correlated, structured data.

[0083] Simultaneously, a complete list of medicinal materials is extracted from multi-party test datasets to construct a comprehensive data list, avoiding the omission of potentially valuable materials. Compatibility support and efficacy support are obtained from the multi-party test datasets, and applicability correlation analysis is performed. Based on the applicability correlation calculation and preset correlation threshold screening, a list of medicinal materials with strong relevance to the target disease and applicable to the disease can be obtained. The quantified applicability correlation improves the comparability of medicinal material evaluation. Finally, a set of candidate medicinal material ratio intervals corresponding to the candidate medicinal material list is constructed. This set of candidate medicinal material ratio intervals ensures quantitative output results and greatly reduces the blind spots in testing.

[0084] Ultimately, finished drugs with efficacy satisfaction below the threshold are screened. Formulation data is extracted to construct a database of undesirable formulations. Reverse screening is then used for efficient risk optimization. The candidate drug formulation range set is then compared with the database of undesirable formulations, eliminating overlapping ranges, mitigating known risks, reducing unsafe ranges, and improving the targeting and success rate of subsequent tests. This results in a reliably optimized set of drug formulation ranges, increasing R&D success rate and reducing risks and costs. Specific optimization suggestions and countermeasures are provided through test data management reports, improving the effectiveness of subsequent research and testing.

[0085] Example 2, as Figure 2 As shown, this application provides a test data management system for innovation in traditional Chinese medicine, including: The demand information receiving module 11 is used to receive research and development demand information of traditional Chinese medicine innovation projects through a user terminal. The research and development demand information includes target diseases and efficacy indicators. The multi-party test data retrieval module 12 is used to connect to the data sharing platform, perform multi-party test data retrieval based on the target disease, and obtain a multi-party test dataset. The correlation analysis module 13 is used to perform medicinal material correlation analysis on the multi-party test dataset, determine the list of candidate medicinal materials that match the target disease, and determine the set of candidate medicinal material ratio intervals; The ratio range optimization module 14 is used to perform constraint optimization on the candidate medicinal material ratio range set based on the efficacy index, and determine the preferred medicinal material ratio range set that meets the efficacy index. The test report generation module 15 is used to generate a test data management report for the traditional Chinese medicine innovation project based on the candidate medicinal material list and the preferred medicinal material ratio range set, and send it to the user terminal.

[0086] In one embodiment, the multi-party test data retrieval module 12 is used for: Connect to a data sharing platform, which includes multiple data storage terminals; Data retrieval conditions are constructed based on the target disease, and data retrieval requests are sent to the multiple data storage terminals based on the data retrieval conditions; Multiple data storage terminals respectively return multiple de-identified datasets based on the data retrieval request; The multiple de-identified datasets are aggregated to obtain the multi-party test dataset.

[0087] The data retrieval conditions for the target disease include: Extract disease feature information of the target disease, including disease classification, pathogenesis, and clinical symptoms; Based on the disease classification, pathogenesis, and clinical symptoms, a search expression is constructed as the data retrieval condition.

[0088] The data retrieval request returns multiple de-identified datasets, including: The first data storage terminal is determined from the plurality of data storage terminals; The first data storage terminal retrieves matching data from the local database according to the data retrieval request to obtain the first matching dataset. The first matching dataset is anonymized to generate a first anonymized dataset, which is then returned. Following the method used to obtain the first de-identified dataset, the other data storage terminals synchronously obtain the corresponding de-identified datasets, resulting in multiple de-identified datasets.

[0089] In one embodiment, the multi-party test data retrieval module 12 is used for: Extract a complete list of medicinal materials from the multi-party test dataset. The complete list of medicinal materials includes multiple Chinese medicines and the range of medicinal material ratios for each Chinese medicine. An applicability correlation analysis is performed on the multiple traditional Chinese medicines and the target disease to obtain multiple applicability correlations. Based on the multiple applicability correlations, the multiple traditional Chinese medicines are screened to obtain a candidate medicinal material list, which includes multiple candidate traditional Chinese medicines. Extract the ratio ranges of multiple candidate Chinese medicinal materials and construct a set of candidate medicinal material ratio ranges corresponding to the candidate medicinal material list.

[0090] The candidate medicinal materials list includes multiple candidate traditional Chinese medicines, including: Extract multiple historical finished drugs from the multi-party test dataset to obtain the total number of finished drugs; The frequency of use of each of the aforementioned traditional Chinese medicines as a preparatory ingredient in the multiple historical finished medicines is statistically analyzed, and the ratio of the frequency of use to the total number of finished medicines is calculated to obtain the compatibility support of each of the aforementioned traditional Chinese medicines. For each of the aforementioned traditional Chinese medicines, the efficacy satisfaction of historical finished products containing the aforementioned traditional Chinese medicines is extracted, and the average of the efficacy satisfaction is calculated to obtain the efficacy support level of each of the aforementioned traditional Chinese medicines. Based on the compatibility support and efficacy support of each of the Chinese herbal medicines, the applicability correlation between each of the Chinese herbal medicines and the target disease is calculated; A preset correlation threshold is set, and Chinese medicines with a correlation greater than the preset correlation threshold are screened to obtain multiple candidate Chinese medicines and generate a candidate medicinal material list.

[0091] In one embodiment, the correlation analysis module 13 is used for: A threshold for treatment satisfaction is set based on the aforementioned treatment indicators; From the historical finished drug data set of the multi-party test dataset, finished drugs with efficacy satisfaction lower than the efficacy satisfaction threshold are selected and identified as the inefficient finished drug set. Extract the proportion data of each candidate Chinese medicine contained in the inefficient finished drug set and establish a database of poor proportions; The candidate medicinal material ratio range set is compared with the undesirable ratio database, and overlapping ratio ranges are eliminated; The eliminated ratio ranges are integrated to generate a set of optimal medicinal material ratio ranges.

[0092] The embodiments of this application, through the specific implementation methods described above, achieve the following technical effects: In this embodiment, the R&D requirements are first concretized into operable technical parameters by the requirement information receiving module 11, providing clear input conditions for subsequent data processing and thus achieving quantitative management of R&D goals. Secondly, the multi-source test data retrieval module 12 connects multiple data storage terminals to build a data sharing platform, providing multi-source data. Disease characteristic information of the target disease is extracted, and a retrieval expression is constructed based on disease classification, pathogenesis, and clinical symptoms, serving as the data retrieval condition. Through the construction of the retrieval expression, a unified expression pattern is formed for intelligent distribution, ensuring data processability. Data retrieval requests are then sent to multiple data storage terminals based on the data retrieval conditions. Accurate execution of the data retrieval requests improves the security of private data.

[0093] Furthermore, the correlation analysis module 13 determines the first data storage endpoint and retrieves matching data from the local database to obtain the first matching dataset. Through anonymization processing, multiple anonymized datasets are obtained, efficiently acquiring data while ensuring its availability and privacy. These anonymized datasets are then integrated to obtain a multi-party test dataset, forming structured data with correlations, which is beneficial for statistical analysis.

[0094] Simultaneously, the ratio interval optimization module 14 extracts a complete list of medicinal materials from the multi-party test dataset, constructing a comprehensive data list to avoid overlooking potentially valuable medicinal materials. The compatibility support and efficacy support are obtained from the multi-party test dataset, and an applicability correlation analysis is performed. Based on the applicability correlation calculation and preset correlation threshold screening, a list of medicinal materials with strong relevance to the target disease and applicable to the disease is obtained, quantifying the applicability correlation and improving the comparability of medicinal material evaluation. Furthermore, a set of candidate medicinal material ratio intervals corresponding to the candidate medicinal material list is constructed to ensure quantitative output results and greatly reduce the blindness of testing.

[0095] Finally, through the test report generation module 15, finished drugs with efficacy satisfaction below the efficacy satisfaction threshold are screened, and formulation data is extracted to construct a database of poor formulations. Reverse screening is then used for efficient risk optimization. The candidate drug formulation range set is then compared with the database of poor formulations to eliminate overlapping ranges, avoid known risks, reduce unsafe ranges, and improve the targeting and success rate of subsequent tests. This results in a reliably optimized set of drug formulation ranges, improving R&D success rate and reducing risks and costs. The test data management report provides specific optimization suggestions and countermeasures to improve the effectiveness of subsequent research and testing.

[0096] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0097] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for managing test data for innovation in traditional Chinese medicine, characterized in that, The method includes: The system receives research and development demand information for innovative traditional Chinese medicine projects through a user terminal. The research and development demand information includes target diseases and efficacy indicators. Connect to the data sharing platform, and retrieve multi-party test data based on the target disease to obtain a multi-party test dataset; Perform medicinal material correlation analysis on the multi-party test dataset to determine a list of candidate medicinal materials that match the target disease, and determine the set of candidate medicinal material ratio intervals; Based on the efficacy indicators, the set of candidate medicinal material ratio intervals is constrained and optimized to determine the set of preferred medicinal material ratio intervals that meet the efficacy indicators. A test data management report for the innovative traditional Chinese medicine project is generated based on the candidate medicinal materials list and the preferred medicinal material ratio range set, and then sent to the user terminal.

2. The method according to claim 1, characterized in that, Connect to the data sharing platform, perform multi-party test data retrieval based on the target disease, and obtain a multi-party test dataset, including: Connect to a data sharing platform, which includes multiple data storage terminals; Data retrieval conditions are constructed based on the target disease, and data retrieval requests are sent to the multiple data storage terminals based on the data retrieval conditions; Multiple data storage terminals respectively return multiple de-identified datasets based on the data retrieval request; The multiple de-identified datasets are aggregated to obtain the multi-party test dataset.

3. The method according to claim 2, characterized in that, Data retrieval criteria are constructed based on the target disease, including: Extract disease feature information of the target disease, including disease classification, pathogenesis, and clinical symptoms; Based on the disease classification, pathogenesis, and clinical symptoms, a search expression is constructed as the data retrieval condition.

4. The method according to claim 2, characterized in that, Multiple data storage terminals respectively return multiple de-identified datasets based on the data retrieval request, including: The first data storage terminal is determined from the plurality of data storage terminals; The first data storage terminal retrieves matching data from the local database according to the data retrieval request to obtain the first matching dataset. The first matching dataset is anonymized to generate a first anonymized dataset, which is then returned. Following the method used to obtain the first de-identified dataset, the other data storage terminals synchronously obtain the corresponding de-identified datasets, resulting in multiple de-identified datasets.

5. The method according to claim 1, characterized in that, Perform medicinal herb association analysis on the multi-party test dataset to determine a list of candidate medicinal herbs matching the target disease, and determine a set of candidate medicinal herb ratio intervals, including: Extract a complete list of medicinal materials from the multi-party test dataset. The complete list of medicinal materials includes multiple Chinese medicines and the range of medicinal material ratios for each Chinese medicine. An applicability correlation analysis is performed on the multiple traditional Chinese medicines and the target disease to obtain multiple applicability correlations. Based on the multiple applicability correlations, the multiple traditional Chinese medicines are screened to obtain a candidate medicinal material list, which includes multiple candidate traditional Chinese medicines. Extract the ratio ranges of multiple candidate Chinese medicinal materials and construct a set of candidate medicinal material ratio ranges corresponding to the candidate medicinal material list.

6. The method according to claim 5, characterized in that, An applicability correlation analysis was performed on the multiple traditional Chinese medicines and the target disease to obtain multiple applicability correlations. Based on these multiple applicability correlations, the multiple traditional Chinese medicines were screened to obtain a candidate medicinal material list. The candidate medicinal material list includes multiple candidate traditional Chinese medicines, including: Extract multiple historical finished drugs from the multi-party test dataset to obtain the total number of finished drugs; The frequency of use of each of the aforementioned traditional Chinese medicines as a preparatory ingredient in the multiple historical finished medicines is statistically analyzed, and the ratio of the frequency of use to the total number of finished medicines is calculated to obtain the compatibility support of each of the aforementioned traditional Chinese medicines. For each of the aforementioned traditional Chinese medicines, the efficacy satisfaction of historical finished products containing the aforementioned traditional Chinese medicines is extracted, and the average of the efficacy satisfaction is calculated to obtain the efficacy support level of each of the aforementioned traditional Chinese medicines. Based on the compatibility support and efficacy support of each of the Chinese herbal medicines, the applicability correlation between each of the Chinese herbal medicines and the target disease is calculated; A preset correlation threshold is set, and Chinese medicines with a correlation greater than the preset correlation threshold are screened to obtain multiple candidate Chinese medicines and generate a candidate medicinal material list.

7. The method according to claim 1, characterized in that, Based on the efficacy indicators, the set of candidate medicinal material ratios is constrained and optimized to determine the preferred set of medicinal material ratios that meet the efficacy indicators, including: A threshold for treatment satisfaction is set based on the aforementioned treatment indicators; From the historical finished drug data set of the multi-party test dataset, finished drugs with efficacy satisfaction lower than the efficacy satisfaction threshold are selected and identified as the inefficient finished drug set. Extract the proportion data of each candidate Chinese medicine contained in the inefficient finished drug set and establish a database of poor proportions; The candidate medicinal material ratio range set is compared with the undesirable ratio database, and overlapping ratio ranges are eliminated; The eliminated ratio ranges are integrated to generate a set of optimal medicinal material ratio ranges.

8. A test data management system for innovation in traditional Chinese medicine, characterized in that, The system for performing the method according to any one of claims 1-7, the system comprising: The demand information receiving module is used to receive research and development demand information of traditional Chinese medicine innovation projects through the user terminal. The research and development demand information includes target diseases and efficacy indicators. The multi-party test data retrieval module is used to connect to the data sharing platform, perform multi-party test data retrieval based on the target disease, and obtain a multi-party test dataset; The correlation analysis module is used to perform medicinal material correlation analysis on the multi-party test dataset, determine the list of candidate medicinal materials that match the target disease, and determine the set of candidate medicinal material ratio intervals; The ratio range optimization module is used to perform constraint optimization on the candidate medicinal material ratio range set based on the efficacy index, and determine the preferred medicinal material ratio range set that meets the efficacy index. The test report generation module is used to generate a test data management report for the traditional Chinese medicine innovation project based on the candidate medicinal material list and the preferred medicinal material ratio range set, and send it to the user terminal.

Images