A method and system for evaluating targeted drug regimen

Abstract

The present invention relates to the field of intelligent medical technology, and in particular to a method and system for evaluating a targeted medication regimen, establishing a comprehensive database, which includes basic information of patients, gene expression data, protein interaction information, and integrated treatment data; obtaining the targeted medication regimen of the current patient; analyzing the gene expression data of the targeted medication regimen, evaluating the efficacy of the targeted medication regimen, and forming an efficacy dimension; constructing a protein interaction network, combining the gene expression data with the protein network, analyzing the molecular characteristics of the patient's tumor, identifying and confirming the drug's target, and forming an individual adaptability dimension; establishing a multidimensional evaluation space, inputting the efficacy dimension and the individual adaptability dimension as independent dimensions, and generating a multidimensional evaluation result output based on the safety dimension and economic dimension of the targeted medication regimen, the efficacy dimension, and the individual adaptability dimension. The problem of the singleness and limitations of the existing targeted medication evaluation method is effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical technology, and in particular to an evaluation method and system for targeted medication regimens. Background Art

[0002] As an important part of precision medicine, targeted drug therapy can significantly improve the specificity and effectiveness of treatment and reduce damage to normal cells by identifying and acting on specific molecular targets. In recent years, with the rapid development of genomics and proteomics technologies, targeted drug therapy has made significant progress in the treatment of complex diseases such as cancer.

[0003] Existing evaluation methods for targeted drug regimens are often limited to single-dimensional data analysis and lack comprehensive consideration of individual patient gene expression and protein interaction information. This limitation leads to incomplete evaluation of targeted drug regimens in multiple aspects, making it difficult to provide accurate predictions of treatment effects and optimization plans.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides a method and system for evaluating a targeted medication regimen, which can effectively solve the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A method and system for evaluating a targeted drug regimen, the method comprising:

[0008] Establishing a comprehensive database, wherein the comprehensive database includes basic information of patients, gene expression data, protein interaction information and integrated treatment data;

[0009] Obtain the current targeted medication regimen for patients;

[0010] Analyzing the gene expression data of the targeted medication regimen based on the comprehensive database, and evaluating the efficacy of the targeted medication regimen to form an efficacy dimension;

[0011] Constructing a protein interaction network, combining the gene expression data with the protein network, analyzing the molecular characteristics of the patient's tumor, identifying and confirming drug targets, and forming an individual adaptability dimension;

[0012] A multidimensional evaluation space is established, and the efficacy dimension and individual adaptability dimension are input as independent dimensions. The safety dimension and economic dimension of the targeted medication regimen are generated according to the comprehensive database, and a multidimensional evaluation result output is generated by integrating the efficacy dimension and individual adaptability dimension.

[0013] Furthermore, the establishment of a comprehensive database includes:

[0014] Enter and store a variety of targeted medication regimen information based on historical medical records;

[0015] Extracting the patient's basic information, gene expression conditions, and protein interaction information from the input data of the multiple targeted medication regimens, and archiving the results in the comprehensive database, and forming a link relationship among the basic information, gene expression data, and protein interaction information under one targeted medication regimen;

[0016] A hierarchical data storage structure is adopted, wherein each layer reflects different types of patient information, and the basic information, gene expression data, protein interaction information and integrated treatment data are stored in different layers respectively;

[0017] Establish a data reprocessing unit to perform data clustering, including data cleaning, standard conversion, and clustering data at the same level;

[0018] A multiple indexing strategy is implemented based on the clustering of the data reprocessing unit, and the sub-data under the multiple targeted medication scheme materials are marked with corresponding timestamps.

[0019] Furthermore, the comprehensive database integrates the patient's basic information, gene expression data, and protein interaction information, and generates an integrated treatment data storage hierarchy, including:

[0020] Creating a multi-dimensional, dynamically updated treatment time series based on targeted medication regimens from different said timestamps;

[0021] Identifying key time points and turning points from the treatment time series, and integrating the basic information, gene expression data, and protein interaction information of the patient according to the key time points and turning points, and obtaining the integrated treatment data;

[0022] The comprehensive database generates an integrated treatment data storage hierarchy for the integrated treatment data, for retrieval of data indexing the timestamps, key time points and turning points.

[0023] Furthermore, the multidimensional evaluation space is established, the efficacy dimension and the individual adaptability dimension are input as independent dimensions, the safety dimension and the economic dimension of the targeted medication scheme are generated according to the comprehensive database, and the multidimensional evaluation result output is generated by combining the efficacy dimension and the individual adaptability dimension, including:

[0024] According to the targeted medication regimen of the current patient, traverse the comprehensive database, extract required data from the comprehensive database, and define evaluation dimensions from various dimensions according to the requirements of the multidimensional evaluation space;

[0025] Selecting an appropriate evaluation algorithm to evaluate the performance of each targeted medication regimen in each evaluation dimension;

[0026] According to different treatment goals and patient needs, weights are set for each evaluation dimension, and each evaluation dimension is weighted to comprehensively generate a multi-dimensional evaluation result output.

[0027] Furthermore, appropriate evaluation algorithms are selected to evaluate the performance of each targeted drug regimen in various evaluation dimensions, including:

[0028] Performing feature indexing on the comprehensive database based on the target targeted medication regimen to obtain a sub-database matching the target targeted medication regimen;

[0029] Standardizing the data of each evaluation dimension in the sub-database to eliminate the dimensional differences between different evaluation dimensions;

[0030] A deep learning method is selected to perform deep learning on the data on each of the evaluation dimensions in the sub-database, and the matching degree between the target targeted medication regimen and the patient's individual biomarkers is evaluated based on the gene expression data and the protein interaction information.

[0031] Furthermore, according to different treatment goals and patient needs, weights are set for each of the evaluation dimensions, and each of the evaluation dimensions is weighted to comprehensively generate a multi-dimensional evaluation result output, including:

[0032] Obtaining the patient's needs, and evaluating the importance of each dimension through an expert system to obtain a preliminary weight of each evaluation dimension;

[0033] Obtain patient data samples with the same needs through the index of the comprehensive database, generate a sample data set from the patient data samples with the same needs, and retrieve the treatment effect corresponding to each sample data in the sample data set;

[0034] Performing deep learning on the sample data set to obtain reference weights on each of the evaluation dimensions;

[0035] According to specific treatment goals, adjusting the preliminary weights according to the reference weights;

[0036] The adjusted weights are used to weight each of the evaluation dimensions to obtain the multi-dimensional evaluation result.

[0037] Furthermore, a training set and a validation set are constructed for the results of the deep learning, and the results of the deep learning are trained and validated respectively.

[0038] Furthermore, the gene expression data of the targeted medication regimen is analyzed based on the comprehensive database, and the efficacy of the targeted medication regimen is evaluated to form an efficacy dimension, including:

[0039] Traversing the gene expression data related to the patient in the comprehensive database, and preprocessing the gene expression data;

[0040] Evaluate the expected efficacy of the targeted drug regimen in the context of different gene expression and quantify the efficacy of the targeted drug on the patient's specific gene expression profile;

[0041] The efficacy dimension is formed by comprehensively considering the regulatory effect of the targeted drug on the target gene, the changes in gene expression levels and the efficacy data in clinical trials.

[0042] Furthermore, the protein interaction network is constructed, the gene expression data is combined with the protein network, the molecular characteristics of the patient's tumor are analyzed, the drug target is identified and confirmed, and the individual adaptability dimension is formed, including:

[0043] Preprocessing the protein interaction data according to the collected protein interaction information;

[0044] Identifying and mapping the positions and relationships of proteins in an interaction network based on integrating the expression data and the protein interaction data of the patient;

[0045] Construct a protein interaction network, where each node represents a protein and each edge represents the interaction between two proteins;

[0046] Performing topological analysis on the protein interaction network and identifying key nodes and key edges in the protein interaction network;

[0047] Extracting protein interaction subnetworks associated with patient tumors from the overall network;

[0048] Analyze the extracted subnetworks, identify key protein and gene pathways, and determine potential drug targets;

[0049] Based on the gene expression data and protein interaction network analysis results, an individual adaptability dimension is formed.

[0050] A method and system for evaluating a targeted drug regimen, the system comprising:

[0051] A comprehensive database module, which establishes a comprehensive database, wherein the comprehensive database includes basic information of patients, gene expression data, protein interaction information and integrated treatment data;

[0052] Get the medication regimen module to obtain the current patient's targeted medication regimen;

[0053] an efficacy dimension analysis module, which analyzes the gene expression data of the targeted medication regimen based on the comprehensive database, and evaluates the efficacy of the targeted medication regimen to form an efficacy dimension;

[0054] A protein interaction network construction module combines the gene expression data with the protein network to analyze the molecular characteristics of the patient's tumor, identify and confirm the drug's target, and form an individual adaptability dimension;

[0055] A multidimensional evaluation module establishes a multidimensional evaluation space, inputs the efficacy dimension and the individual adaptability dimension as independent dimensions, generates the safety dimension and the economic dimension of the targeted medication regimen according to the comprehensive database, and generates a multidimensional evaluation result output in combination with the efficacy dimension and the individual adaptability dimension.

[0056] The technical solution of the present invention can achieve the following technical effects:

[0057] It effectively solves the limitations and singularity of the targeted drug evaluation method under the existing technology, improves the data integration and analysis capabilities, and improves the multi-dimensionality and comprehensiveness of data analysis by establishing a database to integrate the patient's basic information, gene expression data, protein interaction information, and integrated treatment data; it changes the customization method of the treatment plan. The present invention can more accurately identify drug targets and analyze the patient's molecular characteristics, making the treatment plan more personalized, changing the traditional treatment model, and making the treatment more specific and effective; it reduces the side effects and economic burden of treatment. By considering the safety and economy of the treatment plan, the present invention provides doctors and patients with comprehensive evaluation information, reduces damage to normal cells, and considers the patient's economic affordability, thereby reducing the overall treatment burden; it improves the accuracy of the prediction of the treatment effect. The present invention uses advanced algorithms such as deep learning to analyze the rich data in the comprehensive database, and improves the accuracy of the prediction of the treatment effect. This accurate prediction is not only based on statistical data, but also combines the specific biomarkers of the individual, so as to more accurately estimate the efficacy of targeted drugs for specific patients.

[0058] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 The figure is a flowchart of the evaluation method of targeted drug regimen;

[0061] Figure 2 Schematic diagram of the process for establishing a comprehensive database;

[0062] Figure 3 A schematic diagram of the process for establishing a multidimensional evaluation space;

[0063] Figure 4 A flowchart for evaluating the performance of each dimension in the program;

[0064] Figure 5 A flowchart for setting weights for each dimension and outputting comprehensive results;

[0065] Figure 6 Schematic diagram of the process of constructing a protein interaction network;

[0066] Figure 7 Schematic diagram of the structure of the evaluation system for targeted drug regimen. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0069] Embodiment 1

[0070] like Figure 1 As shown, the present application provides a method and a system method for evaluating a targeted drug regimen, the method comprising:

[0071] S100: Establish a comprehensive database that includes basic patient information, gene expression data, protein interaction information, and integrated treatment data;

[0072] S200: Obtain the current targeted medication regimen for the patient;

[0073] S300: Analyze the gene expression data of targeted drug regimens based on a comprehensive database, and evaluate the efficacy of targeted drug regimens to form an efficacy dimension;

[0074] S400: Construct protein interaction networks, combine gene expression data with protein networks, analyze the molecular characteristics of patient tumors, identify and confirm drug targets, and form individual adaptability dimensions;

[0075] S500: Establish a multidimensional evaluation space, input the efficacy dimension and individual adaptability dimension as independent dimensions, generate the safety dimension and economic dimension of the targeted medication scheme based on the comprehensive database, and generate a multidimensional evaluation result output in combination with the efficacy dimension and individual adaptability dimension.

[0076] Specifically, we first create a database to store the patient's basic information, gene expression data, protein interaction information, and integrated treatment data. This database is designed to handle large-scale data sets and support efficient data retrieval and updating. The patient's basic information includes but is not limited to age, gender, medical history, and treatment history. Gene expression data covers the expression levels of various disease-related genes. Protein interaction information includes the interactions and functional relationships between proteins. Next, we obtain the patient's current targeted medication regimen, which is marked as the target regimen for evaluation by the system. We analyze the patient's gene expression data based on statistical and machine learning methods, such as cluster analysis and principal component analysis. Through these analyses, we can predict the effect of drugs on specific The impact of gene expression patterns is used to evaluate the potential efficacy of the treatment plan. Then, a protein network diagram is constructed based on the patient's protein interaction information using the protein database. The network topology is analyzed through network analysis tools to identify key proteins and interaction pathways. An evaluation model is established, and the efficacy dimension and individual adaptability dimension obtained from the previous steps are used as input. In addition, based on the safety and economic data in the database, corresponding evaluation dimensions are generated, and the final multidimensional evaluation results are output. The evaluation results fully consider multiple aspects of targeted medication factors that are adapted to the patient's individual situation. The final output results help doctors understand the advantages and limitations of each plan, so as to make more personalized treatment decisions.

[0077] The technical solution of the present invention effectively solves the limitations and singleness of the targeted drug evaluation method under the existing technology, improves the data integration and analysis capabilities, and improves the multi-dimensionality and comprehensiveness of data analysis by establishing a database to integrate the patient's basic information, gene expression data, protein interaction information and integrated treatment data, changes the customization method of treatment plans, makes treatment plans more personalized, changes the traditional treatment model, and makes the treatment more specific and effective.

[0078] Further, building a comprehensive database, such as Figure 2 As shown, including:

[0079] S110: Input and store multiple targeted medication regimen information based on historical medical records;

[0080] S120: Extract the basic information of patients, gene expression and protein interaction information from the input data of multiple targeted drug regimens, and archive the results in a comprehensive database. The basic information, gene expression data and protein interaction information under a targeted drug regimen form a link relationship.

[0081] S130: A hierarchical data storage structure is adopted, in which each layer reflects different types of patient information, and basic information, gene expression data, protein interaction information, and integrated treatment data are stored in different layers;

[0082] S140: Establishing a data reprocessing unit to perform data clustering processing, including data cleaning, standard conversion, and clustering data at the same level;

[0083] S150: Implement a multiple indexing strategy based on the clustering of the data reprocessing unit, and mark the sub-data under the various targeted medication regimen materials with corresponding timestamps.

[0084] As a preferred embodiment of the above embodiment, first, a comprehensive database is established based on the analysis of historical medical records. The database includes various targeted medication scheme information collected from multiple sources, such as electronic health records, genomic data platforms, and protein interaction databases. The database not only stores the basic information of patients, but also includes detailed gene expression data and protein interaction information. These data will be organized according to different targeted medication schemes to form a link relationship to ensure the integrity and relevance of the information. The data extracted from different medical records are entered into the comprehensive database, including basic information, gene expression data, and protein interaction information. The information under each targeted medication scheme is organized together to form a clear association chain; a hierarchical data storage structure is adopted, and each layer reflects different types of information, such as the first layer stores basic information, the second layer stores gene expression data, the third layer stores protein interaction information, and the fourth layer stores integrated treatment data. This hierarchical storage improves data retrieval efficiency and processing speed; then, a data reprocessing unit is established to be responsible for performing data clustering, data cleaning, standardized conversion, and clustering of data at the same level to ensure data accuracy and consistency, and provide support for analysis and decision-making.

[0085] After clustering, a multi-indexing strategy is implemented. By adding timestamps and performing multiple indexes on the data, the traceability and retrieval of the data are enhanced, which facilitates time series analysis and historical data comparison. Each data item is marked with a corresponding timestamp when it is entered, which helps to track historical changes in the data. Query and analysis can be performed on specific time points or periods. The use of a timestamp-based indexing system ensures that specific data sets can be quickly located when extracting data, thereby improving query efficiency and data processing speed.

[0086] Furthermore, the comprehensive database integrates the patient's basic information, gene expression data, and protein interaction information, and generates an integrated treatment data storage hierarchy, including:

[0087] Create a multi-dimensional, dynamically updated treatment time series based on targeted medication regimens from different timestamps;

[0088] Identify key time points and turning points from the treatment time series, integrate the patient's basic information, gene expression data, and protein interaction information according to the key time points and turning points, and obtain integrated treatment data;

[0089] The comprehensive database generates an integrated treatment data storage hierarchy for integrated treatment data, which is used for data retrieval with indexed timestamps, key time points and turning points.

[0090] As a preferred embodiment of the above, firstly, in the established comprehensive database, the basic information, gene expression data, and protein interaction information of the patient are archived and integrated. In the process of data integration, the integrity and accuracy of the data are maintained, and an integrated treatment data storage hierarchy is created to merge data from different sources into a unified framework to ensure the consistency and availability of the data during the analysis process. Then, based on the targeted medication regimens from different timestamps, a multidimensional and dynamically updated treatment time series is established. This time series will reflect the progress and changes of the patient's treatment, including the adjustment of the treatment regimen and the change of the efficacy. Key time points and turning points are marked for each time point in the time series. The marked key time points and turning points are based on the treatment response or important medical The key time nodes of treatment decision-making are identified through integration of basic information of patients, gene expression data and protein interaction information based on key time points and turning points to capture all important changes related to treatment. The data identified from the treatment time series are integrated to form integrated treatment data, which will be used for further analysis and evaluation. Then, a specific storage level is established in the comprehensive database for the integrated treatment data. A level is set up specifically for storing the integrated data obtained from the treatment time series, which may include but is not limited to treatment effects, drug responses and long-term follow-up health data. An indexing mechanism is set up to quickly retrieve relevant data based on key time points and turning points. The indexing strategy optimizes data retrieval efficiency and improves data management effects.

[0091] Furthermore, a multidimensional evaluation space is established, and the efficacy dimension and individual adaptability dimension are input as independent dimensions. The safety dimension and economic dimension of the targeted medication scheme are generated according to the comprehensive database, and the multidimensional evaluation result output is generated by combining the efficacy dimension and individual adaptability dimension, such as Figure 3 As shown, including:

[0092] S510: according to the current patient's targeted medication regimen, traverse the comprehensive database, extract required data from the comprehensive database, and define evaluation dimensions from various dimensions according to the requirements of the multidimensional evaluation space;

[0093] S520: Select an appropriate evaluation algorithm to evaluate the performance of each targeted drug regimen in each evaluation dimension;

[0094] S530: According to different treatment goals and patient needs, weights are set for each evaluation dimension, and each evaluation dimension is weighted to comprehensively generate a multi-dimensional evaluation result output.

[0095] In this embodiment, first, an evaluation framework is established to process and analyze data from a comprehensive database. The evaluation framework uses the efficacy dimension and the individual adaptability dimension as basic input parameters, and conducts a comprehensive evaluation in combination with the safety and economic dimensions. A multidimensional evaluation model is developed and deployed based on statistical and data analysis tools such as R language or Python. The design of the model will be based on the best practices and proven analysis methods in the medical industry. Then, based on the required data extracted from the comprehensive database, including but not limited to gene expression data, protein interaction information, and the patient's treatment history, different evaluation dimensions are defined according to the requirements of the multidimensional evaluation space. Data mining technology is used to identify and extract data features directly related to efficacy and individual adaptability. Corresponding data is collected for the safety and economic dimensions, such as drug side effect records and treatment costs. Then, a suitable evaluation algorithm is selected to analyze the performance of each targeted drug regimen in different evaluation dimensions, including Machine learning models, such as support vector machines, random forests or neural networks, as well as traditional statistical analysis methods, perform algorithmic evaluations on each dimension and can accurately reflect the characteristics of each dimension and the corresponding treatment effects. Weights are set for each evaluation dimension according to different treatment goals and patient needs. The weights are obtained through expert consultation or historical data analysis to ensure that the importance of each dimension is properly reflected. Each evaluation dimension is weighted according to the set weights to comprehensively generate multi-dimensional evaluation results. The results will help medical professionals understand the advantages and limitations of each targeted medication regimen and provide data support for decision-making. The multi-dimensional evaluation results will be formally output, including but not limited to evaluation reports, charts or real-time data dashboards. These outputs will intuitively display the evaluation results and comprehensive scores of each dimension. Based on the evaluation results, the medical team can optimize the targeted medication regimen and adjust the treatment strategy to meet the specific needs of the patient, while considering the safety and cost-effectiveness of the treatment.

[0096] Furthermore, the performance of each targeted drug regimen in each evaluation dimension is evaluated by selecting an appropriate evaluation algorithm, such as Figure 4 As shown, including:

[0097] S521: feature indexing the comprehensive database based on the target targeted medication regimen to obtain a sub-database matching the target targeted medication regimen;

[0098] S522: Standardizing the data of each evaluation dimension in the sub-database to eliminate the dimensional differences between different evaluation dimensions;

[0099] S523: Select a deep learning method to perform deep learning on the data in each evaluation dimension in the sub-database, based on gene expression data and protein interaction information, and evaluate the matching degree between the target targeted medication regimen and the patient's individual biomarkers.

[0100] As a preferred embodiment of the above, first, a feature indexing system is established to quickly retrieve data matching the current patient's targeted medication regimen from a comprehensive database. The indexing system filters data based on specific parameters of the target targeted medication regimen, such as drug name, target gene, etc.; then, a subset related to a specific targeted medication regimen is extracted from the comprehensive database to create a sub-database, which will be dedicated to analyzing the effects of the specific targeted medication regimen, including gene expression data and protein interaction information; the data in the sub-database is pre-processed, including data cleaning and standardization. The standardization processing methods include but are not limited to normalization and Z-score standardization. The specific method to be selected depends on the distribution characteristics of the data and the evaluation requirements. Data from different sources may have inconsistent formats or different dimensions, so the standardization processing step is necessary to ensure that the comparison between different data dimensions is fair and consistent;

[0101] Then, select an applicable deep learning model to analyze the data in the sub-database. The model can be a convolutional neural network, a recursive neural network, or other advanced models suitable for processing bioinformatics data; train the deep learning model, use the data in the sub-database as a training set, adjust the model parameters to maximize the prediction accuracy and generalization ability, and evaluate the match between the target targeted medication regimen and the patient's individual biomarkers based on the trained deep learning model. The evaluation will be based on the results of the model analysis of gene expression data and protein interaction information; finally, generate an evaluation report that details the interaction and potential efficacy of the drug with the patient's specific biomarkers, as well as any possible side effects or adverse reactions, and integrate the evaluation results of the deep learning model into the decision-making process of the targeted medication regimen. The evaluation results will help the medical team optimize the treatment plan and ensure that the selected drug is most suitable for the patient's specific situation.

[0102] Furthermore, according to different treatment goals and patient needs, weights are set for each evaluation dimension, and each evaluation dimension is weighted to comprehensively generate a multi-dimensional evaluation result output, such as Figure 5 As shown, including:

[0103] S531: Obtain patient needs and evaluate the importance of each dimension through an expert system to obtain the preliminary weight of each evaluation dimension;

[0104] S532: Obtain patient data samples with the same needs through the index of the comprehensive database, generate a sample data set from the patient data samples with the same needs, and retrieve the treatment effect corresponding to each sample data in the sample data set;

[0105] S533: Perform deep learning on the sample data set to obtain reference weights on each evaluation dimension;

[0106] S534: Adjust the preliminary weights according to the reference weights based on the specific treatment goals;

[0107] S535: Use the adjusted weights to weight each evaluation dimension to obtain a multi-dimensional evaluation result.

[0108] As a preferred embodiment of the above, first, detailed patient consultation and demand analysis are carried out. This information will be used to preliminarily set the weights of each evaluation dimension. An expert system or decision support system is used to evaluate the importance of each dimension and generate preliminary weights accordingly. These systems may combine historical data and the latest medical research to provide a scientific basis for setting the weights of each dimension. Next, data samples matching the patient's needs are indexed and extracted from the comprehensive database. The patient data samples with the same needs are organized into a sample data set, and it is ensured that each sample data contains data of each evaluation dimension. These samples include efficacy, safety and economic data under different treatment plans. Deep learning or statistical analysis is performed on the data in the sample data set to obtain reference weights for each evaluation dimension. During the analysis, machine learning algorithms, such as random forests or neural networks, are applied to evaluate the impact of different data points on the treatment outcome. The preliminary weights are adjusted according to the reference weights obtained from the sample data set, and the specific treatment goals and Individual differences among patients are taken into account to ensure that the weights are more in line with the actual treatment effects and patient needs. According to optimization algorithms, such as genetic algorithms or gradient descent methods, the weights are optimized to maximize the accuracy of treatment effect prediction and the relevance of the comprehensive evaluation. The adjusted weights are used to weight each evaluation dimension, and the score of each evaluation dimension is multiplied by the corresponding weight. Then the weighted scores of all dimensions are added together to calculate the comprehensive score of each targeted medication regimen. Different targeted medication regimens are sorted or screened according to the comprehensive score. The higher the comprehensive score, the better the performance of the medication regimen in multiple evaluation dimensions and the more it meets the patient's treatment goals and needs. This step involves merging the scores of each dimension according to the weight to form an overall evaluation result. The result is output as a detailed evaluation report, including the scores of each dimension and the recommended treatment plan. The report will support medical decision-making, help doctors and patients choose the most suitable treatment plan, and adjust the treatment plan according to these results to improve the efficacy and reduce the treatment burden on patients.

[0109] Furthermore, a training set and a validation set are constructed for the results of deep learning to train and validate the results of deep learning respectively.

[0110] Furthermore, the gene expression data of targeted drug regimens are analyzed based on the comprehensive database, and the efficacy of targeted drug regimens is evaluated to form efficacy dimensions, including:

[0111] Traverse the gene expression data related to the patient in the comprehensive database and pre-process the gene expression data;

[0112] Evaluate the expected efficacy of targeted drug regimens in the context of different gene expression and quantify the efficacy of targeted drugs on patient-specific gene expression profiles;

[0113] The efficacy dimension is formed by integrating the regulatory effects of targeted drugs on target genes, changes in gene expression levels and efficacy data in clinical trials.

[0114] In this embodiment, gene expression data in a comprehensive database are traversed. These data include gene expression profiles obtained from patient samples, which may come from a variety of detection platforms. The gene expression data are preprocessed. The cleaning process includes removing noise data and outliers. The standardization process adjusts the proportion of the data so that it can be compared on the same scale. The normalization process ensures that the distribution of the data meets the analysis requirements. Based on the preprocessed gene expression data, a specific algorithm or model is used to evaluate the expected efficacy of the targeted drug regimen in different gene expression backgrounds. The evaluation process takes into account the regulatory effect of the targeted drug on a specific gene and quantifies the effectiveness of the drug in regulating the expression of the target gene. The effect of the targeted drug on the patient's specific gene expression profile is comprehensively analyzed, combined with clinical trial data and literature reports, and the efficacy of the targeted drug is quantified. This includes quantitative analysis of changes in gene expression levels and evaluation of clinical efficacy. The regulatory effect of the targeted drug on the target gene is comprehensively analyzed, specifically including comparing the efficacy differences of the drug in different patient groups and the effectiveness of the drug in a specific gene expression background. Finally, the above analysis results are integrated to generate a quantitative efficacy dimension, which reflects the expected efficacy of the targeted drug in the current patient's specific gene expression background, providing doctors with personalized treatment decision support.

[0115] Further, protein interaction networks are constructed to combine gene expression data with protein networks to analyze the molecular characteristics of patient tumors, identify and confirm drug targets, and form individual adaptive dimensions, such as Figure 6 As shown, including:

[0116] S410: preprocessing the protein interaction data according to the collected protein interaction information;

[0117] S420: Identify and map the positions and relationships of proteins in the interaction network based on the integration of patient expression data and protein interaction data;

[0118] S430: Construct a protein interaction network, where each node represents a protein and each edge represents the interaction between two proteins;

[0119] S440: Perform topological analysis on protein interaction networks and identify key nodes and key edges in protein interaction networks;

[0120] S450: Extracting the protein interaction subnetwork associated with the patient's tumor from the overall network;

[0121] S460: Analyze the extracted subnetworks, identify key protein and gene pathways, and determine potential drug targets;

[0122] S470: Form individual adaptability dimensions based on gene expression data and protein interaction network analysis results.

[0123] In this embodiment, first, protein interaction information related to the target tumor is collected and organized. This information can be used to construct a protein interaction network from a comprehensive database using bioinformatics tools. This network will show the interactions between proteins and mark known and potential drug targets; pre-processing is performed based on the patient's gene expression data, including data standardization and normalization processing to be suitable for subsequent analysis, and the processed gene expression data is combined with the protein interaction network to identify proteins whose expression changes significantly during tumor development and treatment response; then, through topological analysis of the protein interaction network, key nodes and paths in the network are identified. Key nodes may represent important targets in treatment, and key paths may reveal important mechanisms of tumor growth or drug action. According to statistical and network analysis methods, such as network centrality analysis, the importance of these key nodes and paths is identified and evaluated; combining gene expression data and protein interaction network analysis results, biomarker screening methods, such as gene knockout experimental verification and drug sensitivity data, are used to identify and confirm drug targets. In addition, laboratory verification of suspected drug targets can be performed, such as testing the effects of drugs on suspected targets through in vitro cell experiments or in vivo animal models. Protein interaction networks can be continuously updated and optimized based on the latest research findings and clinical trial results. This can include adding new interaction information, adjusting existing connection relationships, and updating node information based on new functional research results. The accuracy and practicality of the network model should be regularly evaluated to ensure that it remains valid in the context of new scientific research and clinical data.

[0124] Embodiment 2

[0125] Based on the same inventive concept as the evaluation system method of a targeted medication regimen in the aforementioned embodiment, the present invention also provides an evaluation method and system for a targeted medication regimen, such as Figure 7 As shown, the system includes:

[0126] Comprehensive database module, to establish a comprehensive database, which includes basic patient information, gene expression data, protein interaction information and integrated treatment data;

[0127] Get the medication regimen module to obtain the current patient's targeted medication regimen;

[0128] The efficacy dimension analysis module analyzes the gene expression data of the targeted drug regimen based on the comprehensive database, and evaluates the efficacy of the targeted drug regimen to form the efficacy dimension;

[0129] The protein interaction network building module combines gene expression data with protein networks to analyze the molecular characteristics of patient tumors, identify and confirm drug targets, and form individual fitness dimensions;

[0130] The multidimensional evaluation module establishes a multidimensional evaluation space, inputs the efficacy dimension and the individual adaptability dimension as independent dimensions, generates the safety dimension and the economic dimension of the targeted medication scheme based on the comprehensive database, and generates a multidimensional evaluation result output in combination with the efficacy dimension and the individual adaptability dimension.

[0131] The above-mentioned adjustment system in the present invention can effectively realize the evaluation method and system of targeted drug regimen, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.

[0132] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for evaluating a targeted drug regimen, characterized in that: The method comprises: Establishing a comprehensive database, wherein the comprehensive database includes basic information of patients, gene expression data, protein interaction information and integrated treatment data; Obtain the current targeted medication regimen for patients; Analyzing the gene expression data of the targeted medication regimen based on the comprehensive database, and evaluating the efficacy of the targeted medication regimen to form an efficacy dimension; Constructing a protein interaction network, combining the gene expression data with the protein interaction network, analyzing the molecular characteristics of the patient's tumor, identifying and confirming drug targets, and forming an individual adaptability dimension; Establish a multidimensional evaluation space, input the efficacy dimension and individual adaptability dimension as independent dimensions, generate the safety dimension and economic dimension of the targeted medication scheme according to the comprehensive database, and generate a multidimensional evaluation result output by integrating the efficacy dimension and individual adaptability dimension, including: According to the targeted medication regimen of the current patient, traverse the comprehensive database, extract required data from the comprehensive database, and define evaluation dimensions from various dimensions according to the requirements of the multidimensional evaluation space; Selecting an appropriate evaluation algorithm to evaluate the performance of each targeted medication regimen in each evaluation dimension; According to different treatment goals and patient needs, weights are set for each evaluation dimension, and each evaluation dimension is weighted to comprehensively generate a multi-dimensional evaluation result output; Select an appropriate evaluation algorithm to evaluate the performance of each targeted drug regimen in each evaluation dimension, including: Performing feature indexing on the comprehensive database based on the target targeted medication regimen to obtain a sub-database matching the target targeted medication regimen; Standardizing the data of each evaluation dimension in the sub-database to eliminate the dimensional differences between different evaluation dimensions; A deep learning method is selected to perform deep learning on the data on each of the evaluation dimensions in the sub-database, and the matching degree between the target targeted medication regimen and the patient's individual biomarkers is evaluated based on the gene expression data and the protein interaction information.

2. The method for evaluating a targeted medication regimen according to claim 1, characterized in that: The establishment of a comprehensive database comprises: Enter and store a variety of targeted medication regimen information based on historical medical records; Extracting the patient's basic information, gene expression conditions, and protein interaction information from the input data of the multiple targeted medication regimens, and archiving the results in the comprehensive database, and forming a link relationship among the basic information, gene expression data, and protein interaction information under one targeted medication regimen; A hierarchical data storage structure is adopted, wherein each layer reflects different types of patient information, and the basic information, gene expression data, protein interaction information and integrated treatment data are stored in different layers respectively; Establish a data reprocessing unit to perform data clustering, including data cleaning, standard conversion, and clustering data at the same level; A multiple indexing strategy is implemented based on the clustering of the data reprocessing unit, and the sub-data under the multiple targeted medication scheme materials are marked with corresponding timestamps.

3. The method for evaluating a targeted medication regimen according to claim 2, characterized in that: The comprehensive database integrates the patient's basic information, gene expression data, and protein interaction information, and generates an integrated treatment data storage hierarchy, including: Creating a multi-dimensional, dynamically updated treatment time series based on the targeted medication regimens from the different timestamps; Identifying key time points and turning points from the treatment time series, and integrating the basic information, gene expression data, and protein interaction information of the patient according to the key time points and turning points, and obtaining the integrated treatment data; The comprehensive database generates an integrated treatment data storage hierarchy for the integrated treatment data, for retrieval of data indexing the timestamps, key time points and turning points.

4. The method for evaluating a targeted medication regimen according to claim 1, characterized in that: According to different treatment goals and patient needs, weights are set for each evaluation dimension, and each evaluation dimension is weighted to comprehensively generate a multi-dimensional evaluation result output, including: Obtaining the patient's needs, and evaluating the importance of each dimension through an expert system to obtain a preliminary weight of each evaluation dimension; Obtain patient data samples with the same needs through the index of the comprehensive database, generate a sample data set from the patient data samples with the same needs, and retrieve the treatment effect corresponding to each sample data in the sample data set; Performing deep learning on the sample data set to obtain reference weights on each of the evaluation dimensions; According to specific treatment goals, the preliminary weights are adjusted according to the reference weights; The adjusted weights are used to weight each of the evaluation dimensions to obtain the multi-dimensional evaluation result.

5. The method for evaluating a targeted medication regimen according to claim 1 or 4, characterized in that: A training set and a validation set are constructed for the results of the deep learning, and the results of the deep learning are trained and validated respectively.

6. The method for evaluating a targeted medication regimen according to claim 1, characterized in that: Analyzing the gene expression data of the targeted medication regimen based on the comprehensive database, and evaluating the efficacy of the targeted medication regimen to form efficacy dimensions, including: Traversing the gene expression data related to the patient in the comprehensive database, and preprocessing the gene expression data; Evaluate the expected efficacy of the targeted drug regimen in the context of different gene expression and quantify the efficacy of the targeted drug on the patient's specific gene expression profile; The efficacy dimension is formed by comprehensively considering the regulatory effect of the targeted drug on the target gene, the changes in gene expression levels and the efficacy data in clinical trials.

7. The method for evaluating a targeted medication regimen according to claim 1, characterized in that: Construct a protein interaction network, combine the gene expression data with the protein interaction network, analyze the molecular characteristics of the patient's tumor, identify and confirm the drug target, and form an individual adaptability dimension, including: Preprocessing the protein interaction network according to the collected protein interaction information; According to integrating the gene expression data of the patient with the protein interaction network, identifying and mapping the positions and relationships of proteins in the interaction network; Construct a protein interaction network, where each node represents a protein and each edge represents the interaction between two proteins; Performing topological analysis on the protein interaction network and identifying key nodes and key edges in the protein interaction network; Extracting protein interaction subnetworks associated with patient tumors from the overall network; Analyze the extracted subnetworks, identify key protein and gene pathways, and determine potential drug targets; Based on the gene expression data and protein interaction network analysis results, an individual adaptability dimension is formed.

8. A targeted drug regimen evaluation system, characterized in that: The method for evaluating a targeted medication regimen according to claim 1 is used, wherein the system comprises: A comprehensive database module, which establishes a comprehensive database, wherein the comprehensive database includes basic information of patients, gene expression data, protein interaction information and integrated treatment data; Get the medication regimen module to obtain the current patient's targeted medication regimen; an efficacy dimension analysis module, which analyzes the gene expression data of the targeted medication regimen based on the comprehensive database, and evaluates the efficacy of the targeted medication regimen to form an efficacy dimension; A protein interaction network construction module combines the gene expression data with the protein interaction network to analyze the molecular characteristics of the patient's tumor, identify and confirm the drug's target, and form an individual adaptability dimension; A multidimensional evaluation module establishes a multidimensional evaluation space, inputs the efficacy dimension and the individual adaptability dimension as independent dimensions, generates the safety dimension and the economic dimension of the targeted medication regimen according to the comprehensive database, and generates a multidimensional evaluation result output in combination with the efficacy dimension and the individual adaptability dimension.

Images