A multi-link intelligent declaration and approval method and system for new traditional Chinese medicine drug research and development

Abstract

The present application relates to the technical field of new drug declaration of traditional Chinese medicine, and particularly relates to a multi-link intelligent declaration and examination and approval method and system for new drug research and development of traditional Chinese medicine, which comprises the following steps: receiving original declaration materials, performing format identification, content analysis, integrity check and format standardization on the original declaration materials to generate a structured declaration document; comparing with current regulations for registration of traditional Chinese medicine to generate a rationality evaluation report of traditional Chinese medicine; performing multi-dimensional technical analysis on the structured declaration document to generate a data integrity report of traditional Chinese medicine; verifying the structured declaration document to generate a data verification report; evaluating the new drug of traditional Chinese medicine to generate a risk report of traditional Chinese medicine; and comprehensively analyzing the results of each evaluation dimension to generate comprehensive evaluation opinions and examination and approval suggestions. The present application improves the transparency, repeatability and efficiency of examination and approval decisions.

Description

Technical Field

[0001] This invention relates to the field of new drug application technology for traditional Chinese medicine, specifically to an intelligent application and approval method and system for multiple stages of new drug research and development for traditional Chinese medicine. Background Technology

[0002] With the continuous improvement of the regulatory system for traditional Chinese medicine (TCM) and the acceleration of its modernization, new TCM drugs must undergo a rigorous application, registration, and review process after their development is completed before they can be used clinically and circulated in the market. This process typically involves multiple stages, including preparation of application materials, regulatory compliance review, technical evaluation, data verification, and comprehensive approval decision-making. It is characterized by a large volume of data, strong interdisciplinary nature, long cycle, and high coordination difficulty.

[0003] Existing patent CN106845943A discloses a pharmaceutical (including medical device) registration management system. Through a registration creation module, a registration process execution guide module, a registration document editing module, and a registration management module, it achieves process-oriented management and progress tracking of registration tasks. However, its core functions remain focused on registration process management and document organization, lacking intelligent analysis, substantive compliance judgment, or scientific evaluation of the technical content of the application materials themselves, and lacking the ability to systematically analyze the authenticity of experimental data, the reliability of clinical data, and potential risks. Existing patent CN113887951B proposes a form-based drug information approval management method, system, and storage medium. Through the construction of approval forms, approval process templates, and business approval flows, it achieves dynamic configuration and optimization of the approval process. However, it focuses on the logical organization and path optimization of the approval process, lacking a deep semantic understanding of technical materials, and does not address key aspects such as multi-source data verification and risk quantification assessment. Existing patent CN112885486A discloses a data-driven pharmacological evaluation and analysis system for traditional Chinese medicine (TCM). This system assesses the pharmacological effects and potential mechanisms of TCM through database construction, active compound screening, target prediction, and compound target network analysis. However, it primarily targets the research and development and pharmacological evaluation stages, focusing its output on pharmacological mechanisms of action and target network analysis. It fails to effectively integrate with drug registration regulations, structured processing of application materials, review process management, data authenticity verification, and approval decision-making mechanisms.

[0004] Therefore, it is necessary to design an intelligent application and approval method and system for multiple stages of new Chinese medicine drug development to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes an intelligent application and approval method and system for multiple stages of new Chinese medicine drug development, aiming to solve the problems of lack of a unified technical framework covering the entire process of new Chinese medicine drug application and registration, insufficient intelligent analysis of application materials, automatic comparison of regulations and technical review support capabilities, and limited means of verifying the authenticity of experimental data and clinical data and quantifying risks.

[0006] On the one hand, this invention proposes an intelligent application and approval method for multiple stages of traditional Chinese medicine new drug development, including: Receive the original application materials, perform format recognition, content parsing, integrity checks and format standardization on the original application materials, and generate a structured application document; The structured application documents are compared with the current regulations for the registration of traditional Chinese medicine to generate a rationality assessment report for traditional Chinese medicine. Multi-dimensional technical analysis is performed on the formulation configuration data, preparation process data, quality control standard data, and pharmacodynamic data in the structured application documents to generate a TCM data integrity report; The test data in the structured application document is verified, and abnormal data is marked and the source tracing analysis process is initiated to generate a data verification report; By integrating the aforementioned TCM data integrity report and data verification report, and combining them with historical adverse drug reaction databases and drug interaction predictions, new TCM drugs are evaluated, and a TCM risk report is generated. A multi-dimensional evaluation matrix is ​​established based on the aforementioned TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report. Through a preset weight allocation algorithm and conflict resolution mechanism, the results of each evaluation dimension are comprehensively analyzed to generate comprehensive review opinions and approval recommendations.

[0007] Furthermore, when generating structured declaration documents, the following are included: Semantic analysis is performed on the original application materials to identify key information fields and establish information relationships; the registration category of new Chinese medicine drugs is identified according to the classification standards for registration of traditional Chinese medicine, and the corresponding application requirements are matched; multi-dimensional checks are performed, including verifying whether the necessary materials are complete, confirming whether the data logic matches, and checking whether the format meets the standard requirements; problematic items are classified and marked, a problem list is generated and modification suggestions are provided; the application content that meets the requirements is converted into a unified format data structure to generate the structured application document.

[0008] Furthermore, when generating a rationality assessment report for traditional Chinese medicine, the following should be included: The structured application documents are assessed for applicability to determine relevant regulations; semantic similarity calculations are used to identify the matching between the application content and the regulatory requirements, quantifying the severity of violations; the identified violations are categorized and prioritized; and a rationality assessment report for traditional Chinese medicine is generated.

[0009] Furthermore, when generating a TCM data integrity report, the following should be included: The traditional Chinese medicine (TCM) theory basis of the formulation was analyzed to confirm that the configuration conforms to the principles of TCM theory; the preparation process was analyzed and verified based on process parameters to identify key process parameters; the matching relationship between pharmacodynamic data and the material basis of TCM was analyzed using pharmacodynamic-component correlation analysis; and a TCM data integrity report was generated.

[0010] Furthermore, when generating a data validation report, the following should be included: The experimental data is verified based on data fingerprints and blockchain evidence; statistical anomaly detection algorithms are used to identify experimental results and clinical data that do not conform to statistical laws; the identified abnormal data is marked and the source tracing analysis process is initiated to track the entire data generation process; and a data verification report is generated.

[0011] Furthermore, the evaluation of new traditional Chinese medicine drugs includes: Based on historical databases of adverse reactions to traditional Chinese medicine and drug interaction knowledge bases, risk prediction models are constructed; the mechanism of action of new traditional Chinese medicine drugs is analyzed based on component-target-disease network analysis to identify potential adverse reactions; clinical trial design and clinical data are evaluated to identify efficacy risks; and quality control systems and testing methods are analyzed to identify quality risks.

[0012] Furthermore, when generating a Traditional Chinese Medicine risk report, the following should be included: The potential adverse reactions, effectiveness risks, and quality risks were scored; the relative weights of each risk factor were determined using the analytic hierarchy process (AHP), and a comprehensive risk index was calculated. Risk levels are generated based on a comprehensive risk index, including low risk, medium risk, and high risk levels; and a structured TCM risk report is generated.

[0013] Furthermore, when establishing a multi-dimensional evaluation matrix, the following should be included: Define a unified indicator system for evaluation dimensions, which includes specified compliance indicators, technical evaluation indicators, data credibility indicators, and risk control indicators; map the information from the TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report to the unified indicator system and perform data normalization processing; Construct a multi-dimensional evaluation matrix, where rows represent evaluation indicators and columns represent evaluation dimensions; assign an initial score to each element in the matrix, and adjust the score based on the correlation between indicators.

[0014] Furthermore, when generating comprehensive review opinions and approval recommendations, the following should be included: Based on different registration categories and drug characteristics, the weights of regulatory compliance, technical evaluation, data credibility, and risk control are dynamically adjusted; when conflicts exist between evaluation dimensions, a conflict resolution mechanism is activated to coordinate through priority rules and expert knowledge base; and comprehensive review opinions and approval recommendations are generated.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By automatically recognizing the format, parsing the semantics, and performing multi-dimensional integrity checks on the original application materials, it can identify problems such as missing materials, logical inconsistencies, and non-standard formats in the early stages of the application process, and provide targeted modification suggestions. This reduces repeated corrections and approval delays caused by problems with the application materials, and improves the first-time pass rate of applications for new traditional Chinese medicine drugs. Based on a structured drug registration regulations knowledge base, the invention automatically compares and reasones the application content according to rules, identifies content items that do not meet the requirements of current regulations, quantifies the degree of violation, and distinguishes between critical and general issues. This reduces compliance risks caused by human misunderstanding and improves the consistency and interpretability of the review results. Simultaneously, it incorporates traditional Chinese medicine theory and modern pharmacology, pharmacology, and quality control theories to conduct multi-dimensional comprehensive analysis of formulation, preparation process, quality standards, and pharmacodynamic data. This reflects both the principles of syndrome differentiation and treatment in traditional Chinese medicine and the requirements of modern drug scientific evaluation, making the technical review more scientific and in line with the technical characteristics of new traditional Chinese medicine drugs. By employing methods such as data fingerprinting, blockchain notarization, and statistical anomaly detection, the authenticity and traceability of experimental and clinical trial data are verified and analyzed, identifying data falsification, alteration, or anomalies, thereby enhancing the credibility of submitted data. Combining historical adverse reaction databases, drug interaction knowledge bases, and component-target-disease network analysis, the safety, efficacy, and quality controllability of new traditional Chinese medicine drugs are comprehensively assessed. A quantitative model outputs risk levels, making risk identification more forward-looking, objective, and quantifiable, facilitating the early identification of potential major risks. By constructing a multi-dimensional evaluation matrix and introducing weighting algorithms and conflict resolution mechanisms, results from multiple aspects, including compliance with regulations, technical scientific validity, data reliability, and risk control, are uniformly integrated and analyzed, reducing subjective human judgment differences and improving the transparency, repeatability, and efficiency of approval decisions.

[0016] On the other hand, this application also provides an intelligent application and approval system for multiple stages of traditional Chinese medicine new drug development, used to apply the above-mentioned intelligent application and approval method for multiple stages of traditional Chinese medicine new drug development, including: The application module is configured to receive original application materials, perform format recognition, content parsing, integrity checks, and format standardization on the original application materials, and generate structured application documents. The specified module is configured to compare the structured application document with the current regulations for the registration of traditional Chinese medicine and generate a rationality assessment report for traditional Chinese medicine. The review module is configured to perform multi-dimensional technical analysis on the formulation configuration data, preparation process data, quality control standard data and pharmacodynamic data in the structured application documents, and generate a TCM data integrity report. The verification module is configured to verify the test data in the structured declaration document, mark abnormal data and initiate the source tracing analysis process, and generate a data verification report. The assessment module is configured to integrate the TCM data integrity report and data verification report, combine them with the historical adverse drug reaction database and drug interaction prediction, assess the new TCM drugs, and generate a TCM risk report. The decision-making module is configured to establish a multi-dimensional evaluation matrix based on the TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report, and to comprehensively analyze the results of each evaluation dimension through a preset weight allocation algorithm and conflict resolution mechanism to generate comprehensive review opinions and approval suggestions. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart of a multi-stage intelligent application and approval method for new traditional Chinese medicine drug development provided in this embodiment of the invention; Figure 2 This is a functional block diagram of the intelligent application and approval system for multi-stage research and development of new traditional Chinese medicine provided in an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] For this, please refer to Figure 1 As shown, this application proposes a multi-stage intelligent application and approval method for new traditional Chinese medicine drug development, including: S100: Receive original application materials, perform format recognition, content parsing, completeness checks, and format standardization on the original application materials, and generate structured application documents; S200: Compare the structured application documents with the current regulations for the registration of traditional Chinese medicine, and generate a rationality assessment report for traditional Chinese medicine; S300: Perform multi-dimensional technical analysis on the formulation configuration data, preparation process data, quality control standard data and pharmacodynamic data in the structured application documents, and generate a TCM data integrity report; S400: Verify the test data in the structured declaration document, mark abnormal data and initiate the source tracing analysis process to generate a data verification report; S500: Integrates TCM data integrity reports and data verification reports, combines historical adverse drug reaction databases and drug interaction predictions to evaluate new TCM drugs and generate TCM risk reports; S600: Establish a multi-dimensional assessment matrix based on the rationality assessment report of traditional Chinese medicine, the data integrity report of traditional Chinese medicine, the data verification report and the risk report of traditional Chinese medicine. Through a preset weight allocation algorithm and conflict resolution mechanism, conduct a comprehensive analysis of the results of each assessment dimension to generate comprehensive review opinions and approval recommendations.

[0020] Specifically, this application proposes a multi-stage intelligent application and approval method for new traditional Chinese medicine (TCM) drug development. This method addresses the entire regulatory process of new TCM drugs, from application acceptance to approval decision-making, and achieves systematic and intelligent processing of application materials, regulatory requirements, technical content, data authenticity, and risk factors, thereby improving the scientific rigor, efficiency, and consistency of TCM drug registration review.

[0021] Specifically, upon receiving the original application materials submitted by the applicant, the system automatically performs format recognition, content parsing, and semantic understanding for documents from different sources and formats, extracting and structuring key information from text, tables, and attachments. Simultaneously, based on the basic requirements for registration of new traditional Chinese medicine drugs, it performs completeness checks and logical consistency verification on the materials, promptly identifying missing items, duplicates, or inconsistencies, and marking and classifying these issues. A unified data model is used to standardize the format of compliant content, generating structured application documents. An internal knowledge base of drug registration regulations is included, covering current regulations for the registration of new traditional Chinese medicine drugs, relevant technical guidelines, and supporting regulatory documents. Through rule-based reasoning and semantic matching technology, the structured application documents are compared item by item with the corresponding regulatory clauses to determine whether the application content meets the requirements and to identify non-compliant or potentially compliance-risk items. For identified non-compliance items, the severity and impact on the approval result are further assessed, generating a traditional Chinese medicine rationality assessment report that includes a description of the violation, legal basis, and rectification suggestions. Integrating traditional Chinese medicine (TCM) theory with modern pharmacology and pharmacology, this system performs multi-dimensional analysis of formulation data, preparation process data, quality control standard data, and pharmacodynamic data in structured application documents. On one hand, it evaluates the theoretical basis and rationality of the formulation from the perspective of TCM syndrome differentiation and treatment and formula compatibility. On the other hand, it analyzes the scientific validity, stability, and reproducibility of the preparation process from the perspective of modern pharmaceutical science. It also evaluates the completeness of quality standards, the applicability of testing methods, and the scientific validity of pharmacodynamic results, ultimately generating a TCM data integrity report. Through secure interface integration with clinical trial registration platforms and trial data management systems, it verifies the source and consistency of data submitted in application documents. Simultaneously, it utilizes technologies such as data fingerprinting, blockchain notarization, and statistical anomaly detection to analyze the originality, completeness, and statistical rationality of experimental and clinical data. For any abnormal data discovered, it automatically marks the data and initiates a traceability analysis process to track the data formation and flow, thereby generating a data verification report. Based on the integration of TCM data integrity reports and data validation reports, and utilizing historical TCM adverse reaction databases and drug interaction prediction models, the safety characteristics, efficacy evidence chains, and quality controllability of new TCM drugs are analyzed. By identifying key risk points such as safety risks, efficacy risks, and quality risks, and quantitatively assessing their probability of occurrence and potential impact, a structured TCM risk report is generated. The output results are integrated and coordinated. Based on the TCM rationality assessment report, TCM data integrity report, data validation report, and TCM risk report, a multi-dimensional assessment matrix is ​​constructed, transforming factors such as regulatory compliance, technical scientific validity, data reliability, and risk control into standardized evaluation indicators.Through a pre-set weighting algorithm and conflict resolution mechanism, the results of each evaluation dimension are comprehensively analyzed and coordinated. When different evaluation conclusions conflict, they are balanced and corrected through priority rules and expert knowledge base, and finally, comprehensive review opinions and approval suggestions are output.

[0022] The working process and principle of this application are as follows: In the initial stage, the original application materials submitted by the applicant are received. These application materials are usually from diverse sources and have different formats, including structured form data as well as unstructured explanatory text, trial reports, and clinical research documents. Format recognition technology is used to identify and classify different types of documents, and natural language processing and semantic parsing algorithms are used to segment, annotate, and extract fields from the text content, establishing the inherent relationships between the application information. Based on this, according to the requirements for registration of new traditional Chinese medicine drugs, the materials undergo multi-dimensional completeness checks and logical consistency verification, identifying missing, conflicting, and non-standard items. Content that meets the requirements is standardized and transformed according to a unified data model, ultimately forming a structured application document. After structuring, using a built-in drug registration regulation knowledge base as the core, a rule-based reasoning mechanism and semantic similarity calculation method are used to automatically compare the application content with current drug registration regulations, technical guidelines, and management methods. First, the specific regulatory clauses applicable to this category of new traditional Chinese medicine (TCM) drug registration are identified. Then, each item in the application is assessed to determine whether it meets the corresponding requirements. Non-compliance items are marked, and the degree and severity of violation are quantified based on their potential impact on the approval results. Through this process, a TCM rationality assessment report is generated, providing objective evidence for subsequent technical review and decision analysis from a regulatory compliance perspective. Subsequently, an in-depth analysis of the technical content of the new TCM drug is conducted. Simultaneously, integrating TCM theoretical knowledge with modern pharmacology, pharmacology, and pharmaceutical engineering, a multi-dimensional analysis model is used to comprehensively review the formulation, preparation process, quality control standards, and pharmacodynamic data in the structured application documents. Its working principle is twofold: firstly, based on TCM syndrome differentiation and treatment and formula compatibility theories, the theoretical basis and rationality of the formulation are evaluated; secondly, from the perspective of modern pharmaceutical science, the scientific validity and reproducibility of process parameters, the completeness of quality standards, and the applicability of testing methods are analyzed. Furthermore, a pharmacodynamic-component correlation model is used to reveal the intrinsic connection between pharmacodynamic results and the material basis of TCM, thereby forming a systematic TCM data integrity report. During or after the technical review process, the authenticity of trial and clinical trial data is independently verified. Through secure interfaces connecting to the clinical trial registration platform and trial data management system, and combining data fingerprinting and blockchain evidence mechanisms, the originality, integrity, and traceability of the data are verified. Simultaneously, statistical anomaly detection algorithms are used to analyze the distribution characteristics of trial results and clinical data, identifying anomalous data deviating from statistical patterns. For detected anomalies, a source tracing analysis process is automatically initiated to track the entire process of data collection, processing, and storage, generating a data verification report accordingly. After obtaining the TCM data integrity report and data verification report, a comprehensive risk assessment of the new TCM drug is conducted.By combining the aforementioned review results with historical databases of adverse reactions to traditional Chinese medicine (TCM) and drug interaction prediction models, and employing a component-target-disease network analysis method, the potential safety risks and adverse reactions of new TCM drugs are assessed. Simultaneously, the completeness of the efficacy evidence chain and the stability of the quality control system are analyzed to identify key risk points. Risk levels are quantified using risk scoring and weighting models, ultimately generating a structured TCM risk report. In the final stage of the approval process, all intermediate results are integrated and comprehensively analyzed. Based on a multi-dimensional evaluation matrix model, information from the TCM rationality assessment report, TCM data integrity report, data validation report, and TCM risk report is mapped to standardized indicators. Data normalization and weighting algorithms are used to comprehensively score different evaluation dimensions. When conflicts arise between different evaluation dimensions, a pre-defined conflict resolution mechanism and an expert knowledge base are used for coordination and correction, ultimately outputting a comprehensive review opinion and approval recommendation.

[0023] As a preferred embodiment, the solution of this application is specifically implemented as follows: The applicant companies submitted a large amount of original materials through the registration platform, including "Pharmaceutical Research Data", "Prescription Composition Description", "Preparation Process Documents", "Quality Standards", "Non-clinical Efficacy and Toxicology Test Reports", and "Phase I-III Clinical Trial Data". The documents were in PDF, Word, Excel and scanned copies.

[0024] The original application materials were subjected to multi-format recognition, utilizing OCR and natural language processing technology to identify scanned text, tables, and images. Subsequently, in accordance with the technical requirements for registration of new traditional Chinese medicine drugs, the materials underwent content parsing and field segmentation, such as automatically extracting the names, dosages, sources, and processing methods of each herb in the prescription, as well as key elements in the process parameters such as temperature, time, and solvent ratio. Based on this, a completeness check was performed, automatically verifying whether any key information was missing (such as whether long-term stability test data was provided, whether the maximum tolerated dose for toxicology was included, etc.), and marking any missing items. Finally, all materials were converted into a unified data model to generate a structured application document.

[0025] After receiving the structured application documents, the system invokes the built-in drug registration regulations knowledge base, which is updated in real time with current regulations and technical specifications such as the "Drug Registration Management Measures," "Classification and Technical Requirements for Traditional Chinese Medicine Registration," and "Technical Guidelines for Clinical Research of New Traditional Chinese Medicine Drugs." In this example, the system automatically compares the application materials with the corresponding requirements for registration of innovative traditional Chinese medicine drugs, such as: whether complete prescription evidence and explanations of the origins of traditional Chinese medicine theory are provided; whether the fingerprint spectrum and multi-component quantitative requirements in the quality standards of traditional Chinese medicine are met; and whether the clinical trial design complies with the requirements for randomization, control, and sample size statistics. When it is found that the description of the control group setting in some clinical trials is insufficient or that some quality standard items do not provide methodological validation data, these are identified as non-compliant items, and a detailed rationality assessment report of traditional Chinese medicine is generated, clearly pointing out the problematic clauses, the relevant legal basis, and rectification suggestions.

[0026] In this example, based on traditional Chinese medicine (TCM) theory, the prescription is analyzed to determine whether its compatibility conforms to theoretical logic and whether there are any drug conflicts or efficacy redundancies. Combining modern pharmacology and pharmaceutical engineering knowledge, a multi-dimensional analysis is conducted on the preparation process data, quality control standards, and pharmacodynamic test results. For example, this includes analyzing whether the extraction process is stable and controllable, and whether key quality attributes (QAs) are correlated with process parameters (CPPs); assessing whether the pharmacodynamic model can reasonably support the proposed indication; and demonstrating the logical consistency between non-clinical safety data and preliminary clinical efficacy. Ultimately, a systematic TCM data integrity report is generated, clearly indicating the advantages and disadvantages of this new TCM drug in terms of scientific rationality, technological maturity, and theoretical support.

[0027] In this example, multi-level verification is performed on the trial and clinical trial data in the application materials. For example, statistical distribution analysis is conducted on non-clinical trial data to detect any abnormal clusters or logical inconsistencies; time series consistency verification is performed on clinical trial data to check whether the case enrollment time, follow-up records, and test results match; and cross-validation is performed with external filing information or ethics approval numbers. When efficacy data from individual centers in a batch of clinical trials is found to deviate significantly from the overall trend, the data is marked as an anomaly, and a source tracing analysis process is initiated to trace back to the specific trial center, original records, and testing personnel, ultimately generating a detailed data verification report.

[0028] In this example, based on the integration of TCM data integrity reports and data validation reports, a historical database of adverse reactions to TCM and a drug interaction prediction model are further utilized to conduct a comprehensive assessment of the drug's safety and efficacy. For example, the incidence of adverse reactions to certain active ingredients in the prescription in previously marketed TCMs is analyzed, and potential toxicity risks are assessed in conjunction with clinical trial results; simultaneously, potential interactions with commonly used drugs are predicted. Finally, safety characteristics, the completeness of the efficacy evidence chain, and quality controllability are quantitatively scored, key risk points are identified, and a TCM risk report with clearly defined risk levels is generated.

[0029] In this example, after receiving the Traditional Chinese Medicine (TCM) rationality assessment report, TCM data integrity report, data validation report, and TCM risk report, a multi-dimensional assessment matrix is ​​constructed. Based on a pre-defined weighting algorithm (e.g., regulatory compliance weighting is higher than process optimization recommendations), and combined with a conflict resolution mechanism (e.g., prioritizing supplementary verification when the technology is acceptable but the data authenticity is questionable), the results of each dimension are comprehensively analyzed. Finally, a comprehensive review opinion is formed, and clear approval recommendations are output, such as "It is recommended to proceed to the next review stage after supplementing some clinical control descriptions and improving the quality standard methodology validation."

[0030] Through the above-mentioned scheme, this application achieves intelligent processing of the entire process of application materials for new traditional Chinese medicine drugs, from receipt, parsing, compliance verification, professional review to comprehensive decision-making. This improves the structure and consistency of application materials, reducing the risk of subjective differences and omissions during manual compilation and review; enhances the accuracy and efficiency of regulatory compliance review and technical review through automatic comparison with drug registration regulations and multi-dimensional review of technical materials; strengthens the traceability and risk controllability of the review process through verification of the authenticity of experimental and clinical data and quantitative risk assessment; and finally, through a multi-dimensional evaluation matrix and weighted decision-making mechanism, forms objective and interpretable comprehensive review opinions and approval recommendations, thereby improving the scientific rigor, standardization, and decision-making efficiency of the registration and approval process for new traditional Chinese medicine drugs.

[0031] This application further proposes methods for generating structured application documents, including: Semantic analysis is performed on the original application materials to identify key information fields and establish information relationships; the registration category of new Chinese medicine drugs is identified according to the classification standards for registration of Chinese medicine, and the corresponding application requirements are matched; multi-dimensional checks are carried out, including verifying whether the necessary materials are complete, confirming whether the data logic matches, and checking whether the format meets the standard requirements; problematic items are classified and marked, a problem list is generated and modification suggestions are provided; and the application content that meets the requirements is converted into a unified format data structure to generate a structured application document.

[0032] Specifically, Natural Language Processing (NLP) technology is used to perform semantic analysis on submitted text materials, tabular data, image scans, and PDF files. This automatically identifies key information fields such as drug name, dosage form, formulation composition, dosage, process parameters, quality standards, experimental data, and clinical trial results, and establishes information relationships between different fields. For example, it matches specific formulation components with corresponding pharmacodynamic data, preparation process parameters, and quality control standards. Then, based on the national classification standards for traditional Chinese medicine registration, it determines whether the new traditional Chinese medicine drug belongs to the category of clinical trial application, marketing authorization application, or generic drug registration, and automatically matches the corresponding application requirements and submission material list. A multi-dimensional completeness check is performed on the application materials, including verifying whether necessary materials are complete, the logical consistency between data items (e.g., whether dosage and pharmacodynamic data match, whether process parameters and quality indicators correspond), and whether the document format conforms to registration specifications (e.g., table templates, units, annotation formats, etc.). Problems found during the check are automatically categorized and marked, such as "missing materials," "data anomalies," or "non-standard format," and a detailed problem list is generated, along with targeted modification suggestions. Finally, all compliant application content is converted into a unified standardized data structure.

[0033] Through the above technical solutions, this application achieves efficient and standardized conversion of raw materials into readable and analyzable data, ensuring the completeness and accuracy of the application materials.

[0034] This application further proposes that, when generating a rationality assessment report for traditional Chinese medicine, the following should be included: The system assesses the applicability of structured application documents and identifies relevant regulations; it uses semantic similarity calculation to identify the match between the application content and the regulatory requirements, quantifying the severity of violations; it categorizes and prioritizes the identified violations; and it generates a Traditional Chinese Medicine (TCM) rationality assessment report, which includes a description of the violations, the regulatory basis, the severity rating of the violations, and modification suggestions.

[0035] Specifically, the system utilizes a built-in drug registration regulations knowledge base. This knowledge base contains the latest regulations, implementation rules, technical guidelines, and various application requirements for new traditional Chinese medicine (TCM) drugs, stored in a structured rule-based format for rapid computer retrieval and logical reasoning. Rule-based reasoning algorithms are used to assess the applicability of various elements in the application documents, such as whether the formulation components, preparation process, quality standards, pharmacodynamic data, and clinical trial design comply with the legal and regulatory requirements for a specific registration category, and automatically match relevant regulations. Subsequently, semantic similarity calculation methods are used to compare the information in the application documents with the regulations in the knowledge base, quantitatively assessing the compliance level and identifying potential violations or inconsistencies. For identified violations, further classification and prioritization are performed, distinguishing between critical violations that may affect drug registration results and general issues. A detailed description is generated for each violation, including the specific violation clause, violation type, potential impact, and suggested modifications. A structured TCM rationality assessment report is automatically generated.

[0036] Through the above technical solution, the generated Traditional Chinese Medicine rationality assessment report not only lists all violations and their severity levels, but also includes clear modification suggestions and referenced legal provisions, enabling the applicant to make targeted supplements or adjustments, thereby improving the compliance and approval rate of the application materials.

[0037] This application further proposes that when generating a Traditional Chinese Medicine data integrity report, the following should be included: The TCM theory basis of the formulation was analyzed to confirm that the configuration conforms to the principles of TCM theory; the preparation process was analyzed and verified based on process parameters to identify key process parameters; the matching relationship between pharmacodynamic data and the material basis of TCM was analyzed using pharmacodynamic-component correlation analysis; and a TCM data integrity report was generated, which includes technical advantages, existing problems, and improvement suggestions.

[0038] Specifically, the process begins by utilizing the built-in TCM theory knowledge base and modern pharmacology database to analyze the formulation data of new TCM drugs. This involves evaluating the TCM theoretical basis of the formulation, such as the rationality of the compatibility of the medicinal materials, whether the properties and meridian tropism of the herbs conform to the principles of TCM syndrome differentiation and treatment, and the scientific validity of the overall treatment approach. Simultaneously, modern pharmacology is used to analyze the chemical composition of the active ingredients in the formulation, assessing their efficacy potential and safety. For the preparation process data, process parameter analysis is used to evaluate the scientific validity, repeatability, and operational controllability of the preparation method, identifying critical process parameters (CPPs) and their impact on the final drug quality. Regarding quality control standards, the testing indicators in the application documents are compared and analyzed with current pharmacopoeias, technical specifications, and industry standards to assess their coverage, applicability, and sensitivity, ensuring the quality control system is complete, operable, and reliable. Furthermore, pharmacodynamic-component correlation technology is employed to match and analyze the experimental pharmacodynamic data with the material basis of traditional Chinese medicine, identifying the correlation and mechanism of action between active ingredients and pharmacological effects, thus providing support for the scientific validity of pharmacodynamics. Based on the above analysis, a data integrity report for traditional Chinese medicine is generated, comprehensively listing the technical advantages of the proposed new traditional Chinese medicine (such as innovative formulation, advanced preparation process, and improved quality control), existing problems (such as unclear key process parameters and insufficient pharmacodynamic correlation of some active ingredients), and improvement suggestions for addressing these problems (such as optimizing process parameters, supplementing pharmacodynamic verification experiments, and improving quality testing methods).

[0039] Through the above technical solutions, this application ensures that the prescription design conforms to the traditional Chinese medicine theory and treatment principles while taking into account modern pharmacological logic by evaluating the rationality of the formulation; through the analysis of preparation process parameters and the identification of key processes, it ensures the scientific nature, reproducibility, and controllability of the drug production process; the comparative analysis of quality control standards can verify the applicability and completeness of the detection methods, ensuring the controllability of drug quality; and combined with the pharmacodynamic-component correlation analysis, it can clarify the correspondence between active ingredients and pharmacological effects, improving the scientific nature of pharmacodynamic verification.

[0040] This application further proposes that when generating a data verification report, the following should be included: The experimental data is verified based on data fingerprints and blockchain evidence; statistical anomaly detection algorithms are used to identify experimental results and clinical data that do not conform to statistical laws; the identified abnormal data is marked and the source tracing analysis process is initiated to track the entire data generation process; a data verification report is generated, which includes data authenticity rating, a list of abnormal data, source tracing analysis results, and verification conclusions.

[0041] Specifically, the system interfaces with external data sources such as clinical trial registration platforms, electronic data management systems, and laboratory information management systems through secure authentication interfaces to verify the availability and accessibility of registration numbers, ethics approval numbers, trial center lists, and original data files. Subsequently, it generates data fingerprints for the received data (e.g., calculating SHA-256 checksums for original files and key data tables, constructing a Merkle tree and recording the Merkle root), and associates these fingerprints with blockchain-based evidence records (including timestamps, submitter public key signatures, and transaction hashes) to prove the data's originality and that the submission time has not been tampered with. In parallel, it performs multi-dimensional statistical and semantic consistency checks on the data, including range checks, missing value pattern analysis, time series consistency verification, inter-center / inter-subject variability comparisons, and randomization and blinding integrity checks. It employs classical statistical methods (such as Z-scores, t-tests, and ANOVA), distribution tests (e.g., Benford's rule for detecting numerical distribution anomalies), and modern anomaly detection algorithms (such as Isolation Forest, Locality of Anomalies (LOF), and time series-based anomaly detection models) to identify records that do not conform to statistical regularities or abnormal patterns. Detected anomalies are automatically categorized (e.g., data entry errors, instrument calibration issues, center offset, suspicious signs of fraud, protocol deviation, etc.) and prioritized according to their impact. For high-priority anomalies, a source tracing analysis process is automatically initiated, tracking the data generation chain along metadata and audit logs—including the mapping between sample ID and testing center, testing instrument and batch number, testing time, data entry account and electronic signature, original paper records or source instrument documents. Automated queries or evidence collection tasks can be initiated with the testing center (e.g., requesting the upload of original QC documents, calibration certificates, and laboratory operator signature records). Throughout the process, the complete audit trail is recorded and an evidence package is generated (including blockchain transaction IDs, fingerprint comparison results, snapshots of differences between raw / processed data, and audit log fragments). Finally, based on a rule engine and risk scoring model, the verified data is given a data authenticity rating (e.g., "high credibility / medium credibility / low credibility" or quantified on a scale of 0–100), and a detailed data verification report is output. The structured report includes: an overall authenticity rating and confidence level description; a list of anomalous data (each item listing the type of anomalous data, affected data rows / subjects / centers, statistical indicators, and a summary of evidence); source tracing analysis results and preliminary root cause assessments (including found evidence and missing items); blockchain / fingerprint verification proof (including transaction hashes and timestamps); specific recommendations and priorities for further verification or rectification (e.g., requesting the submission of original records, recalibrating instruments, or conducting audits of specific centers); a machine-readable summary of key points (JSON format) for direct access by the review and decision-making modules, along with a link to an evidence package available for manual review.

[0042] Through the aforementioned technical solutions, this application connects with the clinical trial registration platform and trial data management platform via a secure authentication interface, achieving traceability of data sources and controllable access, ensuring that submitted data is original and legal. Utilizing data fingerprints and blockchain notarization, the integrity and immutability of trial data are encrypted and verified, forming an auditable chain of evidence. Through statistical anomaly detection algorithms, abnormal data is automatically identified and marked, and source tracing analysis is initiated to track the entire data generation process, preventing data falsification or entry errors. The generated data verification report provides a comprehensive data authenticity rating, a list of abnormal data, source tracing analysis results, and verification conclusions.

[0043] This application further proposes that the evaluation of new traditional Chinese medicine drugs include: Based on historical databases of adverse reactions to traditional Chinese medicine and drug interaction knowledge bases, risk prediction models are constructed; the mechanism of action of new traditional Chinese medicine drugs is analyzed based on component-target-disease network analysis to identify potential adverse reactions; clinical trial design and clinical data are evaluated to identify efficacy risks; and quality control systems and testing methods are analyzed to identify quality risks.

[0044] Specifically, the study retrieves historical adverse reaction databases for traditional Chinese medicine (TCM), including adverse reaction reports, dosages, patient group characteristics, and adverse reaction incidence rates for various TCMs and their compound preparations. It also accesses a drug interaction knowledge base to obtain potential interaction information for different drug combinations. Based on this data, a risk prediction model is constructed. This model, based on a Bayesian network, uses nodes to represent drug components, targets, and adverse reaction events, and edges to represent the conditional probability relationships between components, targets, and adverse reactions, thereby predicting the types and probabilities of potential adverse reactions for new TCM drugs. Simultaneously, the model is trained on historical data using machine learning algorithms (such as random forests or gradient boosting trees) to quantify the risk weights of different component combinations, achieving accurate prediction of potential safety risks. A scientific analysis of the clinical trial design is conducted, including sample size calculation, randomization methods, control group design, endpoint selection, and follow-up time arrangements. Finally, by combining pharmacodynamic data and experimental data from the TCM data integrity report, the efficacy evidence chain of the new TCM drug is evaluated. The construction of this efficacy evidence chain is based on a causal inference model. By establishing a multi-layered correlation network of "component-target-disease-efficacy," it quantifies the contribution of each pharmacodynamic component to clinical efficacy, identifies missing evidence links, and predicts efficacy changes under different usage conditions, thereby ensuring logical consistency between clinical data and the pharmacological basis. The preparation process parameters, key quality control indicators, and detection methods are analyzed. Based on statistical process control (SPC) and multivariate statistical process control (MSPC) models, the controllability of the preparation process and quality indicators is analyzed, identifying key process parameters and risk factors affecting batch-to-batch consistency. Through quantitative assessment of the integrity of the quality control system, the reliability of detection methods, and process repeatability, improvement suggestions are provided, and a quality risk level is output.

[0045] Through the above technical solutions, this application integrates the integrity report and data verification report of traditional Chinese medicine data, and combines the historical database of adverse reactions to traditional Chinese medicine and the knowledge base of drug interactions to conduct a comprehensive and multi-dimensional assessment of the safety, efficacy and quality controllability of new traditional Chinese medicine drugs. This enables the quantitative prediction of potential safety risks, the complete construction of the evidence chain of efficacy, and the reliability analysis of the quality control system, thereby identifying key risk points and proposing optimization suggestions.

[0046] This application further proposes that when generating a Traditional Chinese Medicine risk report, the following should be included: Potential adverse reactions, efficacy risks, and quality risks are scored; the relative weights of each risk factor are determined using the analytic hierarchy process (AHP), and a comprehensive risk index is calculated. Risk levels are generated based on a comprehensive risk index, including low risk, medium risk, and high risk levels; and a structured TCM risk report is generated.

[0047] Specifically, an in-depth analysis of the safety, efficacy, and quality control information of new traditional Chinese medicine drugs is conducted, and potential risk points are categorized and organized. Safety risks include potential adverse reactions, toxicity risks, and drug interactions; efficacy risks include insufficient clinical efficacy, incomplete evidence chains, or flawed trial design; quality risks include raw material fluctuations, deviations from key process parameters, imperfect testing methods, or poor batch-to-batch consistency. For each category of key risk points, the probability of occurrence and potential severity of each risk point are calculated by combining historical adverse reaction databases, drug interaction knowledge bases, and experimental and clinical data. Subsequently, an Analytic Hierarchy Process (AHP) is used to construct a risk assessment model, allocating weights to various risk factors according to their impact on the overall drug safety, efficacy, and quality, thereby calculating a comprehensive risk index. The comprehensive risk index reflects the relative importance and potential harm of a single risk point within the overall risk structure. Based on a pre-set risk grading standard, the comprehensive risk index is mapped to three levels: low risk, medium risk, and high risk, to clarify the urgency and priority of each risk point. Furthermore, the classification information, scoring results, weight calculations, comprehensive risk index, and corresponding risk levels of all key risk points are compiled to generate a structured TCM risk report.

[0048] Through the above technical solutions, this application not only presents each risk point and its quantitative indicators in the form of tables, graphs and scoring matrices, but also provides improvement suggestions and control measures for each risk point, thereby improving the efficiency and accuracy of risk identification in the registration and approval process of new Chinese medicine drugs.

[0049] This application further proposes the following for establishing a multi-dimensional evaluation matrix: Define a unified indicator system for evaluation dimensions. The unified indicator system includes specified compliance indicators, technical evaluation indicators, data credibility indicators, and risk control indicators. Map the information from the TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report to the unified indicator system and perform data normalization processing.

[0050] Specifically, in establishing a multi-dimensional evaluation matrix based on the TCM rationality assessment report, TCM data integrity report, data validation report, and TCM risk report, a standardized indicator system was first defined to ensure the comparability of various evaluation results under a unified scale. This standardized indicator system comprises four core dimensions: regulatory compliance indicators, technical and scientific indicators, data reliability indicators, and risk control indicators. Regulatory compliance indicators primarily assess the degree of compliance between the TCM new drug application materials and current drug registration regulations, including the applicability and matching of various regulatory clauses, the severity of violations, and the implementation of proposed amendments. Technical and scientific indicators primarily evaluate the rationality of the formulation, the scientific nature and reproducibility of the preparation process, the completeness of quality control standards, the applicability of testing methods, and the correlation between pharmacodynamic data and the material basis of TCM. Data reliability indicators assess the authenticity, completeness, and statistical consistency of experimental and clinical trial data, including anomaly labeling, source tracing analysis results, and validation conclusions. Risk control indicators measure the effective management of identified key risk points and their overall risk levels. When mapping information from various assessment reports to a standardized indicator system, all scores, qualitative descriptions, and data entries in the original reports undergo unified transformation and normalization. For example, the severity of violations, the number of technical issues, the probability of data anomalies, and risk indices are quantified into standardized values ​​ranging from 0 to 1; qualitative descriptions such as "high risk" and "needs improvement" are converted into corresponding standardized scores. This normalization process ensures that data from different sources and in different formats can be directly compared and comprehensively calculated on the same dimension, avoiding biases caused by differences in the scale or expression of the original data.

[0051] Through the above technical solution, each element in the matrix of this application contains a normalized score and indicator weight, which not only realizes the systematic integration of various evaluation reports, but also provides quantitative, traceable and efficient technical support for the scientific decision-making of registration and approval of new Chinese medicine drugs.

[0052] This application further proposes that, when establishing a multi-dimensional evaluation matrix, the following should also be included: Construct a multi-dimensional evaluation matrix, where rows represent evaluation indicators and columns represent evaluation dimensions; assign an initial score to each element in the matrix, and adjust the score based on the correlation between indicators.

[0053] Specifically, the matrix is ​​first defined by its rows and columns: the rows represent various evaluation indicators, such as regulatory compliance coverage, formulation rationality score, reproducibility of the preparation process, completeness of quality control standards, reliability of pharmacodynamic data, authenticity of experimental data, degree of identification of key risk points, and comprehensive risk index; the columns correspond to the four core evaluation dimensions, including regulatory compliance, technical scientific validity, data reliability, and risk control. After the matrix is ​​constructed, an initial score is assigned to each element based on the quantitative data, qualitative scores, and normalization results from each report. For example, for the "formulation rationality score" indicator, the initial score may come from a comprehensive evaluation of traditional Chinese medicine theory and modern pharmacological basis; for the "authenticity of experimental data" indicator, the initial score comes from an assessment of the authenticity, completeness, and abnormal detection results of experimental and clinical trial data; for the "regulatory compliance coverage" score, the score comes from an analysis of the matching degree between the structured application documents and drug registration regulations and the severity of violations. After the initial score allocation, the scores are also corrected based on the correlation between the indicators. Scoring adjustments include the following aspects: First, adjusting scores based on the logical dependencies between indicators; for example, uncontrollable key process parameters can affect the quality control standard score. Second, considering the impact of cross-dimensional factors; for example, data reliability issues may lower the credibility score of the technical and scientific dimension. Third, introducing historical review data and experience weights to weight certain key indicators to reflect actual approval risks and key concerns. Scoring adjustments typically employ algorithmic models, such as weighted averages, analytic hierarchy process (AHP), or Bayesian network methods, to ensure that the scores in the matrix not only reflect the results of individual indicators but also the mutual influence and overall consistency between indicators.

[0054] Through the above technical solution, this application generates an accurate and comprehensive multi-dimensional evaluation matrix, which not only quantifies various evaluation results, but also visualizes the correlation between various indicators.

[0055] This application further proposes methods for generating comprehensive review opinions and approval recommendations, including: Based on different registration categories and drug characteristics, the weights of regulatory compliance, technical evaluation, data credibility, and risk control are dynamically adjusted; when conflicts exist between evaluation dimensions, a conflict resolution mechanism is activated to coordinate through priority rules and expert knowledge base; and comprehensive review opinions and approval recommendations are generated.

[0056] Specifically, the weights of the four core assessment dimensions—regulatory compliance, technical scientific validity, data reliability, and risk control—are dynamically adjusted based on the registration category of the new traditional Chinese medicine (TCM) (such as chemically derived TCM, compound TCM, and single-ingredient preparations) and drug characteristics (such as new indications, complexity of compound formulations, and clinical urgency). For example, for new drug applications or TCM compound preparations involving high-risk ingredients, the weights of risk control and data reliability may be significantly increased. Conversely, for drugs with similar approval experience, the weights of regulatory compliance and technical scientific validity may be higher, reflecting the differentiated focus of the approval process. During the comprehensive analysis, if there are assessment conflicts between different dimensions—for example, if the technical scientific validity assessment shows a reasonable formulation and sufficient efficacy, but the data reveals abnormalities or suspicious trends in clinical trial data—a conflict resolution mechanism will be activated. This mechanism includes the following steps: first, determining the relative importance of each dimension in affecting the approval result based on preset priority rules; second, accessing the expert knowledge base and comprehensively considering historical approval cases and the experience of domain experts to correct and weight the conflicting issues; and finally, generating a reconciled score or grade to ensure that the conflict does not affect the overall assessment logic. The conflict resolution mechanism can also highlight abnormal data or high-risk points to ensure that approvers can focus their review. After comprehensive analysis, the weighted scores of each dimension are summarized to form a final comprehensive evaluation value of a multi-dimensional assessment matrix. This value is then combined with regulatory requirements and historical approval experience to generate a comprehensive review opinion. This opinion not only includes an overall evaluation of the safety, efficacy, quality controllability, and risk level of the new traditional Chinese medicine drug, but also specifically lists key findings, potential problems, improvement suggestions, and priority approval recommendations for each dimension. A structured approval report can be output, including dimensional scores, a comprehensive score, conflict resolution records, and expert opinion citations.

[0057] Through the above technical solutions, this application has achieved intelligent management of the entire process from multi-dimensional data integration, weight allocation, conflict handling to the generation of final approval recommendations, which improves the scientific nature, efficiency and consistency of the registration and approval of new Chinese medicine drugs, reduces human subjective judgment errors, and ensures that key risk points are identified and assessed.

[0058] This application also provides an intelligent application and approval system for multiple stages of traditional Chinese medicine new drug development, used to apply the above-mentioned intelligent application and approval method for multiple stages of traditional Chinese medicine new drug development, including: The application module is configured to receive original application materials, perform format recognition, content parsing, integrity checks, and format standardization on the original application materials, and generate structured application documents. The specified module is configured to compare structured application documents with current regulations for the registration of traditional Chinese medicine and generate a rationality assessment report for traditional Chinese medicine. The review module is configured to perform multi-dimensional technical analysis on the formulation configuration data, preparation process data, quality control standard data and pharmacodynamic data in the structured application documents, and generate a TCM data integrity report. The verification module is configured to verify the test data in the structured declaration document, mark abnormal data and initiate the source analysis process to generate a data verification report; The assessment module is configured to integrate TCM data integrity reports and data verification reports, combine them with historical adverse drug reaction databases and drug interaction predictions, assess new TCM drugs, and generate TCM risk reports. The decision-making module is configured to establish a multi-dimensional evaluation matrix based on the TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report. It then uses a preset weighting algorithm and conflict resolution mechanism to comprehensively analyze the results of each evaluation dimension and generate comprehensive review opinions and approval recommendations.

[0059] In summary, by automatically identifying the format, parsing the semantics, and performing multi-dimensional integrity checks on the original application materials, issues such as missing materials, logical inconsistencies, and non-standard formats can be identified early in the application process. Targeted modification suggestions are then provided, reducing repeated revisions and approval delays caused by problems with the application materials, and increasing the first-time pass rate for new traditional Chinese medicine (TCM) drug applications. Based on a structured drug registration regulations knowledge base, automatic comparison and rule-based reasoning of the application content identify items that do not comply with current regulations, quantify the degree of violation, and distinguish between critical and general issues. This reduces compliance risks caused by human misunderstanding and improves the consistency and interpretability of the review results. Furthermore, by incorporating TCM theoretical knowledge and modern pharmacology, pharmacology, and quality control theories, a multi-dimensional comprehensive analysis of formulation, preparation process, quality standards, and pharmacodynamic data is conducted. This approach reflects both the TCM principles of syndrome differentiation and treatment and the formulation principles, while also meeting the requirements of modern drug scientific evaluation, making the technical review more scientific and consistent with the technical characteristics of new TCM drugs. By employing methods such as data fingerprinting, blockchain notarization, and statistical anomaly detection, the authenticity and traceability of experimental and clinical trial data are verified and analyzed, identifying data falsification, alteration, or anomalies, thereby enhancing the credibility of submitted data. Combining historical adverse reaction databases, drug interaction knowledge bases, and component-target-disease network analysis, the safety, efficacy, and quality controllability of new traditional Chinese medicine drugs are comprehensively assessed. A quantitative model outputs risk levels, making risk identification more forward-looking, objective, and quantifiable, facilitating the early identification of potential major risks. By constructing a multi-dimensional evaluation matrix and introducing weighting algorithms and conflict resolution mechanisms, results from multiple aspects, including compliance with regulations, technical scientific validity, data reliability, and risk control, are uniformly integrated and analyzed, reducing subjective human judgment differences and improving the transparency, repeatability, and efficiency of approval decisions.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A multi-stage intelligent application and approval method for the research and development of new traditional Chinese medicine drugs, characterized in that, include: Receive the original application materials, perform format recognition, content parsing, integrity checks and format standardization on the original application materials, and generate a structured application document; The structured application documents are compared with the current regulations for the registration of traditional Chinese medicine to generate a rationality assessment report for traditional Chinese medicine. Multi-dimensional technical analysis is performed on the formulation configuration data, preparation process data, quality control standard data, and pharmacodynamic data in the structured application documents to generate a TCM data integrity report; The test data in the structured application document is verified, and abnormal data is marked and the source tracing analysis process is initiated to generate a data verification report; By integrating the aforementioned TCM data integrity report and data verification report, and combining them with historical adverse drug reaction databases and drug interaction predictions, new TCM drugs are evaluated, and a TCM risk report is generated. A multi-dimensional evaluation matrix is ​​established based on the aforementioned TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report. Through a preset weight allocation algorithm and conflict resolution mechanism, the results of each evaluation dimension are comprehensively analyzed to generate comprehensive review opinions and approval recommendations.

2. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 1, characterized in that, When generating structured application documents, the following should be included: Semantic analysis is performed on the original application materials to identify key information fields and establish information relationships; the registration category of new Chinese medicine drugs is identified according to the classification standards for registration of traditional Chinese medicine, and the corresponding application requirements are matched; multi-dimensional checks are performed, including verifying whether the necessary materials are complete, confirming whether the data logic matches, and checking whether the format meets the standard requirements; problematic items are classified and marked, a problem list is generated and modification suggestions are provided; the application content that meets the requirements is converted into a unified format data structure to generate the structured application document.

3. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 2, characterized in that, When generating a rationality assessment report for traditional Chinese medicine, the following should be included: The structured application documents are assessed for applicability to determine relevant regulations; semantic similarity calculations are used to identify the matching between the application content and the regulatory requirements, quantifying the severity of violations; the identified violations are categorized and prioritized; and a rationality assessment report for traditional Chinese medicine is generated.

4. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 3, characterized in that, When generating a Traditional Chinese Medicine data integrity report, the following should be included: The traditional Chinese medicine (TCM) theory basis of the formulation was analyzed to confirm that the configuration conforms to the principles of TCM theory; the preparation process was analyzed and verified based on process parameters to identify key process parameters; the matching relationship between pharmacodynamic data and the material basis of TCM was analyzed using pharmacodynamic-component correlation analysis; and a TCM data integrity report was generated.

5. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 4, characterized in that, When generating a data validation report, the following should be included: The experimental data is verified based on data fingerprints and blockchain evidence; statistical anomaly detection algorithms are used to identify experimental results and clinical data that do not conform to statistical laws; the identified abnormal data is marked and the source tracing analysis process is initiated to track the entire data generation process; and a data verification report is generated.

6. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 5, characterized in that, The evaluation of new traditional Chinese medicine drugs includes: Based on historical databases of adverse reactions to traditional Chinese medicine and drug interaction knowledge bases, risk prediction models are constructed; the mechanism of action of new traditional Chinese medicine drugs is analyzed based on component-target-disease network analysis to identify potential adverse reactions; clinical trial design and clinical data are evaluated to identify efficacy risks; and quality control systems and testing methods are analyzed to identify quality risks.

7. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 6, characterized in that, When generating a Traditional Chinese Medicine risk report, the following should be included: The potential adverse reactions, effectiveness risks, and quality risks were scored; the relative weights of each risk factor were determined using the analytic hierarchy process (AHP), and a comprehensive risk index was calculated. Risk levels are generated based on a comprehensive risk index, including low risk, medium risk, and high risk levels; and a structured TCM risk report is generated.

8. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 7, characterized in that, When establishing a multi-dimensional evaluation matrix, the following should be included: Define a unified indicator system for evaluation dimensions, which includes specified compliance indicators, technical evaluation indicators, data credibility indicators, and risk control indicators; map the information from the TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report to the unified indicator system and perform data normalization processing; Construct a multi-dimensional evaluation matrix, where rows represent evaluation indicators and columns represent evaluation dimensions; assign an initial score to each element in the matrix, and adjust the score based on the correlation between indicators.

9. The intelligent application and approval method for multi-stage research and development of new traditional Chinese medicine drugs according to claim 8, characterized in that, When generating comprehensive review opinions and approval recommendations, the following should be included: Based on different registration categories and drug characteristics, the weights of regulatory compliance, technical evaluation, data credibility, and risk control are dynamically adjusted; when conflicts exist between evaluation dimensions, a conflict resolution mechanism is activated to coordinate through priority rules and expert knowledge base; and comprehensive review opinions and approval recommendations are generated.

10. A multi-stage intelligent application and approval system for the research and development of new traditional Chinese medicine drugs, used to apply the multi-stage intelligent application and approval method for the research and development of new traditional Chinese medicine drugs as described in any one of claims 1-9, characterized in that, include: The application module is configured to receive original application materials, perform format recognition, content parsing, integrity checks, and format standardization on the original application materials, and generate structured application documents. The specified module is configured to compare the structured application document with the current regulations for the registration of traditional Chinese medicine and generate a rationality assessment report for traditional Chinese medicine. The review module is configured to perform multi-dimensional technical analysis on the formulation configuration data, preparation process data, quality control standard data and pharmacodynamic data in the structured application documents, and generate a TCM data integrity report. The verification module is configured to verify the test data in the structured declaration document, mark abnormal data and initiate the source tracing analysis process, and generate a data verification report. The assessment module is configured to integrate the TCM data integrity report and data verification report, combine them with the historical adverse drug reaction database and drug interaction prediction, assess the new TCM drugs, and generate a TCM risk report. The decision-making module is configured to establish a multi-dimensional evaluation matrix based on the TCM rationality assessment report, TCM data integrity report, data verification report, and TCM risk report, and to comprehensively analyze the results of each evaluation dimension through a preset weight allocation algorithm and conflict resolution mechanism to generate comprehensive review opinions and approval suggestions.

Images