Multi-center data processing method, device, electronic device and medium in clinical trials

Abstract

The present application discloses a multi-center data processing method, device, electronic device and medium in a clinical trial, which belongs to the field of data processing technology, wherein the multi-center data processing method in a clinical trial includes: obtaining multi-source clinical data of a clinical trial; determining the data type of each data in the multi-source clinical data, and dynamically adjusting the weight of each data source according to the characteristics of the data type to obtain a first weight; determining the clinical characteristics of individual patients in the multi-source clinical data, and dynamically adjusting the weighting rules according to the clinical characteristics to obtain the weighting rules; processing the multi-source clinical data according to the first weight and the weighting rule. The method adjusts the weight of each variable in real time according to the different characteristics of the data set and the actual needs of clinical analysis, so that the degree of influence of different characteristics on the final analysis results is reasonably distributed. The dynamic update and optimization of weights are realized, which significantly improves the accuracy and adaptability of data analysis.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and specifically relates to a multi-center data processing method, device, electronic equipment and medium in a clinical trial. Background Art

[0002] With the rapid development of modern medical technology and the acceleration of medical informatization, the types and quantities of medical data have exploded. Clinical data comes from various sources, including patients' electronic medical records (EHR), laboratory test data, medical images, genomic data, and case data from multi-center clinical trials. These data play a vital role in medical diagnosis, treatment optimization, disease prevention, and medical research.

[0003] However, the complexity of these data also brings great challenges. There are significant differences in data collection standards and quality among different medical centers, which may lead to deviations in analysis results. Summary of the invention

[0004] The purpose of this application is to provide a multi-center data processing method, device, electronic device and medium in a clinical trial to update and optimize weights in real time using a dynamic weighting mechanism, thereby improving the accuracy of data analysis.

[0005] According to a first aspect of an embodiment of the present application, a method for processing multi-center data in a clinical trial is provided, and the method may include: Access multi-source clinical data for clinical trials; Determine the data type of each data in the multi-source clinical data, and dynamically adjust the weight of each data source according to the characteristics of the data type to obtain a first weight; Determine the clinical characteristics of individual patients in multi-source clinical data, and dynamically adjust the weighting rules according to the clinical characteristics to obtain the weighting rules; Multi-source clinical data are processed according to first weights and weighting rules.

[0006] In some optional embodiments of the present application, after processing the multi-source clinical data according to the first weight and the weighting rule, the multi-center data processing method in the clinical trial further includes: The first weight is optimized by using machine learning and adaptive optimization algorithm to obtain a second weight; The multi-source clinical data are processed according to the second weights and weighting rules.

[0007] In some optional embodiments of the present application, the first weight is optimized by using a machine learning and adaptive optimization algorithm to obtain a second weight, including: A multi-dimensional prediction model was constructed based on historical clinical data, and random forests were used to analyze the contribution of different features to clinical outcomes; The prediction model is trained using the multi-source clinical data processed according to the first weight and the weighting rule, so that the prediction model improves the prediction accuracy.

[0008] In some optional embodiments of the present application, training the prediction model using multi-source clinical data processed according to the first weight and weighting rule includes: The entropy weight method and Bayesian optimization method are used to optimize the weighting coefficients of each data source in multi-source clinical data.

[0009] In some optional embodiments of the present application, before processing the multi-source clinical data according to the second weight and weighting rule, the multi-center data processing method in the clinical trial further includes: Conduct consistency analysis on multi-source clinical data to obtain analysis results; Based on the analysis results, the multi-source clinical data were adjusted for consistency to obtain adjusted data.

[0010] In some optional embodiments of the present application, after the consistency adjustment of the multi-source clinical data is performed according to the analysis results, the multi-center data processing method in the clinical trial further includes: The adjusted data were validated for consistency using cross-validation and independent validation set methods to ensure data consistency.

[0011] In some optional embodiments of the present application, after processing the multi-source clinical data according to the second weight and the weighting rule, the multi-center data processing method in the clinical trial further includes: Obtain the latest clinical data of patients; Adjust the second weight, weighting rules and analysis results according to the latest clinical data.

[0012] According to a second aspect of an embodiment of the present application, a multi-center data processing device in a clinical trial is provided, and the device may include: The acquisition module is used to obtain multi-source clinical data for clinical trials; A first determination module is used to determine the data type of each data in the multi-source clinical data, and dynamically adjust the weight of each data source according to the characteristics of the data type to obtain a first weight; The second determination module is used to determine the clinical characteristics of individual patients in the multi-source clinical data, and dynamically adjust the weighting rules according to the clinical characteristics to obtain a first weighting rule; A processing module is used to process multi-source clinical data according to a first weight and a first weighting rule.

[0013] According to a third aspect of an embodiment of the present application, an electronic device is provided, which may include: processor; a memory for storing processor-executable instructions; The processor is configured to execute instructions to implement the multi-center data processing method in a clinical trial as shown in any one of the embodiments of the first aspect.

[0014] According to the fourth aspect of the embodiments of the present application, a storage medium is provided. When the instructions in the storage medium are executed by a processor of an information processing device or a server, the information processing device or the server implements a multi-center data processing method in a clinical trial as shown in any one of the embodiments of the first aspect.

[0015] The above technical solution of the present application has the following beneficial technical effects: The method of the embodiment of the present application adjusts the weight of each variable in real time according to the different characteristics of the data set and the actual needs of clinical analysis, so that the influence of different characteristics on the final analysis results is reasonably distributed, and the dynamic update and optimization of weights are realized, which significantly improves the accuracy and adaptability of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] FIG1 is a schematic flow chart of a multi-center data processing method in a clinical trial in an exemplary embodiment of the present application; FIG2 is a schematic diagram of the structure of a multi-center data processing device in a clinical trial in an exemplary embodiment of the present application; FIG3 is a schematic diagram of the structure of an electronic device in an exemplary embodiment of the present application; Figure 4 It is a schematic diagram of the hardware structure of an electronic device in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with specific implementations and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present application. In addition, in the following description, the description of known structures and technologies is omitted to avoid unnecessary confusion of the concepts of the present application.

[0018] The accompanying drawings show schematic diagrams of layer structures according to embodiments of the present application. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clarity. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0019] Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0020] In the description of the present application, it should be noted that the terms "first", "second" and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0021] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0022] In conjunction with the accompanying drawings, the multi-center data processing method, device, electronic device and medium in the clinical trial provided by the embodiments of the present application are described in detail through specific embodiments and their application scenarios.

[0023] like Figure 1 As shown, in a first aspect of an embodiment of the present application, a method for processing multi-center data in a clinical trial is provided, and the method may include: S110: Obtain multi-source clinical data for clinical trials; S120: Determine the data type of each data in the multi-source clinical data, and dynamically adjust the weight of each data source according to the characteristics of the data type to obtain a first weight; S130: determining clinical characteristics of individual patients in multi-source clinical data, and dynamically adjusting weighting rules according to the clinical characteristics to obtain weighting rules; S140: Processing multi-source clinical data according to the first weight and weighting rule.

[0024] The embodiment method of the present application adjusts the weight of each variable in real time according to the different characteristics of the data set and the actual needs of clinical analysis, so that the influence of different characteristics on the final analysis results is reasonably distributed. Dynamic updating and optimization of weights are achieved, and the accuracy and adaptability of data analysis are significantly improved. Among them, the weighted rule is a mechanism for dynamically adjusting data weights based on the individual clinical characteristics of patients. Through this mechanism, different data can play a more reasonable role in the analysis, improve the accuracy and reliability of the analysis results, and provide more powerful support for clinical decision-making. It can include weight adjustment rules based on data characteristics and weight adjustment rules combined with individual patient characteristics.

[0025] Among them, the weight adjustment rules based on data characteristics: a multi-factor adaptive algorithm is used to dynamically adjust the weight of each data source according to the characteristics of different data types (data scarcity, volatility, data quality, etc.), and the entropy weight method is combined with the Bayesian update mechanism to evaluate the real-time weight of each type of data; the weight adjustment rules combined with the individual characteristics of patients: according to the individual clinical characteristics of patients (age, gender, medical history, etc.), a multi-model integration method is used to train exclusive weighted models for different disease groups, and the weighting rules are further dynamically adjusted.

[0026] In this embodiment, the entropy weight method can be used to perform preliminary weight allocation on each data source. The entropy weight method is an objective weight determination method that automatically determines the weight of each data source by measuring the amount of information and uncertainty of each data source. The specific steps are as follows: Data standardization: Standardize the features of all data sources to eliminate the impact of different dimensions and ensure that all features are comparable.

[0027] Information entropy calculation: Evaluate the information entropy of each data source. The higher the information entropy, the more dispersed the information of the data source, the greater the uncertainty, and the lower the corresponding weight. Conversely, data sources with lower information entropy have higher weights because they provide more valuable information.

[0028] Weight determination: Based on the calculated information entropy, the preliminary weight of each data source is assigned. In this way, data sources with large amounts of information receive higher weights, and data sources with smaller amounts of information receive lower weights, ensuring the objectivity and rationality of the preliminary weight assignment.

[0029] After completing the initial weight assignment, the Bayesian optimization method can be introduced to further refine and optimize the weights. Bayesian optimization is an optimization technique based on a probability model that can efficiently search for the best weight combination to improve the model's predictive performance. The specific steps are as follows: Define optimization goals: Set optimization goals, such as maximizing the accuracy of the prediction model or minimizing the prediction error. These goals drive the Bayesian optimization process to find the optimal weight combination.

[0030] Surrogate model construction: Build a surrogate model to approximate the changes in model performance under different weight combinations. This surrogate model can be constructed using methods such as Gaussian processes to efficiently predict model performance during the optimization process.

[0031] Iterative optimization process: Sampling and Evaluation: New weight combinations are selected for evaluation based on the surrogate model. The weight combinations selected are usually those that are expected to significantly improve model performance.

[0032] Performance evaluation: Apply the selected weight combination to the data sources, weightedly integrate the numerical data, and evaluate its predictive performance after training the model.

[0033] Model update: Based on the evaluation results, the proxy model is updated and the selection strategy of future weight combinations is adjusted. This process is continuously iterated to gradually approach the optimal weight combination.

[0034] Determine the optimal weights: After multiple iterations, identify the optimal weight combination that maximizes the model's prediction accuracy or minimizes the prediction error. These weights ensure that the contribution of each data source in the comprehensive analysis is the most accurate and effective.

[0035] In some embodiments, after processing the multi-source clinical data according to the first weight and the weighting rule, the method for processing multi-center data in clinical trials further includes: The first weight is optimized by using machine learning and adaptive optimization algorithm to obtain a second weight; The multi-source clinical data are processed according to the second weights and weighting rules.

[0036] This embodiment can adjust the weight distribution by introducing a weight optimization algorithm based on the entropy weight method and Bayesian optimization, maximize the accuracy and reliability of data processing, solve the subjective bias and error problems in traditional methods, and ensure the objectivity and optimality of weight distribution.

[0037] After completing the weight optimization, the optimized weights are applied to the numerical data of each data source for weighted integration. The specific steps include: Weighted integration: According to the optimized weights, the numerical data of each data source are weighted and integrated to generate a comprehensive data set. This process ensures that the contribution of each data source in the comprehensive analysis meets the optimization goal.

[0038] Model validation: Retrain the prediction model using the weighted synthetic dataset and validate its performance on an independent test set. Confirm the effectiveness of weight optimization by comparing model performance indicators before and after optimization.

[0039] Continuous monitoring and updating: The system regularly monitors changes in model performance and data characteristics, reapplies entropy weighting and Bayesian optimization methods when necessary, and dynamically adjusts weights to maintain optimal model performance.

[0040] In some embodiments, the first weight is optimized using a machine learning and adaptive optimization algorithm to obtain a second weight, including: A multi-dimensional prediction model was constructed based on historical clinical data, and random forests were used to analyze the contribution of different features to clinical outcomes; The prediction model is trained using the multi-source clinical data processed according to the first weight and the weighting rule, so that the prediction model improves the prediction accuracy.

[0041] In some embodiments, training a prediction model using multi-source clinical data processed according to the first weight and the weighting rule includes: The entropy weight method and Bayesian optimization method are used to optimize the weighting coefficients of each data source in multi-source clinical data.

[0042] During the model training process, the weights are dynamically updated and the optimization strategy is adjusted in real time in combination with the feedback of real-time clinical data. During a specific treatment process, some data sources (drug dosage, laboratory test results) may have a greater impact on clinical decision-making, while some data sources (imaging data) may not be critical for the time being. Through the real-time learning mechanism, the weights can be continuously adjusted according to the latest data to adapt to the dynamic changes in the treatment process.

[0043] Feedback adjustment formula:

[0044] in, Represents the amount of data correction from real-time feedback, ensuring that weight adjustments are not only based on prediction errors but also take into account actual clinical feedback.

[0045] :For i The dynamic weight of the class data at time t+1 is the weight value after an update based on the original time t. As time goes by and the data changes, the weight will be continuously adjusted to reflect the reasonable degree of influence of different data on the analysis results at different times.

[0046] :It is i The dynamic weight of the class data at time t is the weight value at the previous moment. It is the basis for calculating the weight at the next moment, reflecting the continuity of the weight and the starting point of dynamic updating.

[0047] γ: is the adjustment coefficient, which is mainly used to control the amplitude of weight adjustment. It determines the degree of influence of the real-time feedback data correction on the weight adjustment. If γ is large, the weight adjustment amplitude is relatively large, and the model responds to new data more quickly; if it is small, the weight adjustment amplitude is small, and the model changes relatively more smoothly, which can avoid excessive weight adjustment due to data fluctuations. In some embodiments, before processing multi-source clinical data according to the second weight and weighting rule, the multi-center data processing method in clinical trials also includes: Conduct consistency analysis on multi-source clinical data to obtain analysis results; Based on the analysis results, the multi-source clinical data were adjusted for consistency to obtain adjusted data.

[0048] In the process of combining multidimensional data sources, there are often data consistency issues, that is, there are errors or information conflicts between different data sources. In order to ensure the accuracy and consistency of the analysis results, this technical solution adopts consistency detection and correction methods to ensure the coordination between data sources. The specific steps are as follows: Step 1: Data consistency check By designing a consistency evaluation function, the differences between data sources are calculated, and statistical methods are used to evaluate the matching degree of different data sources (laboratory data and imaging data) in the same clinical scenario.

[0049] Step 2: Data correction and fusion According to the consistency analysis results, weighted correction technology is used to adjust inconsistent data. The weighted average method is used to correct the differences between data.

[0050] Step 3: Model validation and clinical feedback Through model validation and clinical feedback, we ensured that the final prediction model had high consistency and stability in different patient groups. We used cross-validation and independent validation set methods to evaluate the performance of the model on different data sets to ensure consistency.

[0051] In some embodiments, after the consistency adjustment of the multi-source clinical data is performed according to the analysis results, the multi-center data processing method in the clinical trial further includes: The adjusted data were validated for consistency using cross-validation and independent validation set methods to ensure data consistency.

[0052] Consistency verification formula: .

[0053] Among them, k is the number of validation sets For the validation set, For the test set.

[0054] This embodiment can ensure that the results after each data update can maintain a high degree of consistency through chi-square test and weighted fusion method, combined with multiple iterations and cross-validation. Even if the data set changes, the algorithm can still ensure the stability of the analysis results through consistency evaluation and weight adjustment, avoiding decision errors caused by data fluctuations.

[0055] In some embodiments, after processing the multi-source clinical data according to the second weight and the weighting rule, the method for processing multi-center data in clinical trials further includes: Obtain the latest clinical data of patients; Adjust the second weight, weighting rules and analysis results according to the latest clinical data.

[0056] Based on the results of the above dynamic weighting, weight optimization and consistency analysis, a clinical decision support system is finally formed. The system can automatically adjust the analysis model according to the patient's clinical data in real time.

[0057] Through the interface, it can connect with the hospital's electronic medical record (EMR) system, laboratory information management system (LIS) and other data sources in real time to obtain the patient's latest clinical data. This process can be achieved through the API interface and data synchronization module to ensure the real-time and integrity of the data.

[0058] The system uses an optimized weighted model to analyze data and combine clinical history data and treatment plans for intelligent recommendations. Specifically, the system uses the aforementioned dynamic weighting mechanism and optimization algorithm to process the input data and generate prediction results and recommended plans.

[0059] The prediction formula can be:

[0060] Among them, y is the prediction result; is the standardized value of the i-th feature; It is a summation symbol, indicating that the cumulative calculation is performed from i=1 to i=n, ​​and n represents the number of features involved in the calculation, which means that several different features are considered in the model to obtain the final prediction result, covering the multi-source clinical data characteristics of the patient; : is the dynamic weight of the ith feature at time t. This weight reflects the relative importance of the ith feature to the prediction result at the current moment. For example, when analyzing the condition of a heart disease patient, if the change in blood pressure data during the current treatment stage is critical to the condition judgment, then the blood pressure-related feature corresponding to The weight will be higher; f i (D t ,p): is the characteristic function of the i-th type of data, which is used to describe the relationship between the data characteristics and the prediction results of the i-th type of data at time t and the specific situation of the patient individual p. This function will vary according to different data types and analysis purposes. Its main function is to convert the original data into a form that is meaningful for prediction. Taking blood glucose data as an example, (D t ,p) may transform blood glucose measurements based on factors such as the patient's diabetes type, treatment stage, etc. to accurately reflect its contribution to the prediction of the disease.

[0061] Based on the recommended results output by the model, doctors can make more scientific and personalized clinical decisions, improve treatment effects and reduce misdiagnosis rates. The system displays prediction results and recommended plans through the user interface, including risk assessment, treatment recommendations, monitoring indicators, etc., to assist doctors in making decisions.

[0062] For example, the user interface displays the following content: Risk Assessment Chart: Displays the probability distribution of a patient's risk of developing a disease.

[0063] Recommended treatment plan list: recommended treatment plans based on the predicted results and their advantages and disadvantages.

[0064] Monitoring indicator dynamic graph: real-time display of changing trends of key health indicators.

[0065] Historical data comparison: Compare current data with historical data to show changing trends.

[0066] This embodiment aims to build a multi-dimensional prediction model based on historical clinical data, using a random forest algorithm to analyze the contribution of different features to clinical outcomes. The specific implementation process is as follows: 1. Data Collection and Preparation Data collection: Collect historical clinical numerical data from multiple medical information systems (such as electronic medical records EMR, laboratory information system LIS, etc.). The data includes demographic information (such as age, gender), medical history, treatment records, and clinical outcomes (such as cure rate, survival rate).

[0067] Data cleaning: Use appropriate methods (such as mean filling or median filling) to fill in missing data. Identify and correct or delete obviously erroneous data points to ensure data accuracy and consistency. Standardize numerical features (such as normalization or Z-Score standardization) to eliminate the impact of different dimensions and ensure the effectiveness of model training.

[0068] 2. Feature Selection Key feature identification: Select key features that are closely related to clinical outcomes from the collected data, such as blood sugar levels, blood pressure, electrocardiogram data, etc.

[0069] Correlation analysis: Statistical methods are used to evaluate the correlation between each feature and the clinical outcome, and to screen out features that have a significant impact on the prediction results.

[0070] 3. Model Training Random forest model construction: Use the selected key features as input to build a random forest model. Random forest can effectively capture the complex relationship between features and improve the accuracy of prediction by integrating multiple decision trees.

[0071] Model training: The dataset is divided into a training set and a test set (e.g., 70% for training and 30% for testing). A random forest model is trained on the training set to learn how to predict clinical outcomes based on the input features.

[0072] 4. Feature Importance Analysis Evaluate feature contribution: Evaluate the contribution of each feature to the prediction result through the feature importance score of the random forest model. Identify the features that have the greatest impact on clinical outcomes and provide a basis for subsequent weighted adjustments.

[0073] During the data collection process, the present application also provides that the present technical solution uses a series of innovative methods to efficiently and intelligently collect, process and integrate multi-source numerical clinical data, ensuring the comprehensiveness, accuracy and high quality of the data, and providing a solid data foundation for subsequent dynamic weighting, weight optimization and consistency analysis.

[0074] By designing a dynamic data format recognition and unified conversion process, it is possible to automatically recognize the numerical data formats from different medical information systems and convert them into a unified internal standard format.

[0075] Specific steps: Data format scanning: The system automatically scans the input numerical data and identifies its source system and format type (such as CSV, JSON, XML, etc.).

[0076] Format mapping and conversion: Based on predefined format mapping rules, the system automatically converts the identified numerical data into a unified internal standard format. For example, laboratory test results in CSV format and real-time monitoring data in JSON format are converted into unified database entries.

[0077] Automatic verification: After conversion, the system automatically verifies the consistency and integrity of the data to ensure accuracy during the conversion process. Through checksum or data integrity verification mechanisms, it ensures that the converted data conforms to the predetermined format and standards.

[0078] By introducing a real-time protocol monitoring and automatic adjustment mechanism, the data adaptation process can be automatically adjusted when the medical information system protocol changes, ensuring the continuity and stability of data collection.

[0079] Specific steps: Protocol monitoring: The system continuously monitors changes in data transmission protocols of various medical information systems and captures protocol changes in real time by subscribing to protocol update notifications or regular protocol scanning.

[0080] Automatic update: When a protocol change is detected, the system automatically updates the data conversion rules and adaptation process. For example, the rule engine can dynamically load new conversion templates or parameters without manual intervention.

[0081] Seamless switching: Ensure that data collection is not interrupted during the protocol update process and maintain the continuity of data flow. Ensure a smooth transition between the old and new protocols through version control and transaction management.

[0082] A real-time numerical data preprocessing process based on distributed processing can be developed, which can efficiently process large-scale numerical data and ensure the quality and consistency of the data before entering the integration stage.

[0083] Specifically, the data processing process may include: Data cleaning: Detect and remove noise and erroneous values ​​in numerical data in real time, and use rule-based filtering and simple statistical methods (such as Z-Score to detect outliers) for preliminary cleaning.

[0084] Data aggregation: According to predefined rules, real-time data is aggregated to generate useful statistical indicators. For example, the real-time blood pressure data is averaged at the minute level to reduce the amount of data while retaining key trend information.

[0085] Quality Control: Perform a series of quality control checks to ensure the accuracy and integrity of the data, including data consistency checks, range verification, and logic verification to ensure that the pre-processed data meets the analysis requirements.

[0086] A dynamic data collection mechanism based on event triggering can be constructed, which can automatically start or pause data collection tasks according to specific conditions, thereby improving the efficiency and response speed of data collection.

[0087] Specific steps: Event recognition: The system sets key statistical thresholds or patterns, for example, triggering an abnormal event when blood glucose levels exceed 180 mg / dL or fall below 70 mg / dL for five consecutive minutes.

[0088] Task Scheduling: Automatically start or adjust data collection tasks based on triggered events. For example, increase the sampling frequency to obtain more detailed monitoring data, or focus on specific data categories to deeply analyze the causes of abnormalities.

[0089] Real-time response: Ensure that the data acquisition system can respond quickly when an event occurs, and obtain and process key numerical data in a timely manner. Ensure the efficient execution of key tasks through priority scheduling and resource allocation optimization.

[0090] A context-aware numerical data cleaning process can be developed during data processing and integration, which can intelligently identify and correct outliers and missing values ​​based on the patient's historical data and clinical background.

[0091] Specific steps: Historical data analysis: The system analyzes the patient's historical numerical data to establish normal ranges and trend models. For example, the normal range of blood sugar fluctuations for patients can be determined through time series analysis.

[0092] Anomaly detection: Based on the established model, outliers and missing values ​​in the current data are automatically identified. Abnormal fluctuations are detected using moving average or trend analysis methods.

[0093] Intelligent correction: Automatically correct or complete detected problematic data based on contextual information and historical trends. For example, use linear interpolation or nearest neighbor-based data to fill missing values ​​to ensure data continuity and accuracy.

[0094] A dynamically adjusted numerical standardization process can be designed, which can automatically adjust the standardization parameters according to different data sources and clinical needs to ensure the comparability and consistency of numerical data on a unified scale.

[0095] Specific steps: Standardization parameter setting: The system presets different standardization parameters (such as mean, standard deviation) based on the data source and clinical needs. For example, blood glucose measurement equipment in different hospitals may have different measurement standards.

[0096] Real-time adjustment: Automatically adjust the normalization parameters based on the real-time monitored data distribution to ensure data consistency on a uniform scale. Dynamically calculate normalization parameters through sliding window technology to adapt to the latest distribution changes of data.

[0097] Unified output: The standardized numerical data is output to a unified data warehouse for subsequent analysis. This ensures that all data is stored and processed under the same standards, improving data comparability and analysis accuracy.

[0098] A multi-source numerical data fusion process can be established to efficiently integrate numerical data from different systems and build a unified patient numerical view.

[0099] Specific steps: Data matching: Matching numerical data from different data sources through unique identifiers (such as patient ID) to ensure that data from different sources for the same patient can be correctly associated.

[0100] Data fusion: Use weighted average or rule-based fusion methods to integrate matching data into a unified numerical view. For example, if a certain indicator comes from multiple data sources, the system performs a weighted average based on the reliability of the data source and the timestamp of the data to generate a comprehensive value.

[0101] Consistency check: Perform consistency check on the fused data to ensure the accuracy and completeness of the data. Logical verification and range verification are performed to ensure that the integrated data meets the expected clinical standards.

[0102] During the implementation process, an intelligent conflict detection and resolution process can be introduced to automatically identify and resolve data conflicts and inconsistencies in the process of integrating multi-source numerical data.

[0103] Specific steps: Scan the integrated numerical data to identify possible conflicts and inconsistencies (such as different values ​​for the same indicator).

[0104] Conflicts should be categorized based on data source and clinical importance, for example, data from high-reliability data sources should be prioritized, or the most recent data should be selected based on the chronological order of the data.

[0105] According to predefined rules or context information, the most reliable data source is automatically selected or a comprehensive value is generated through interpolation to resolve data conflicts. For example, if there are differences in blood pressure values ​​measured by different devices, the system can select the most reliable value or calculate the average value as the final blood pressure value based on the patient's historical blood pressure trend and current measurement conditions.

[0106] Automatically identify multi-source numerical data formats and use deep generative models to achieve unified conversion, reduce manual intervention, and improve data compatibility. Through adaptive control systems, respond to protocol changes in real time to ensure a stable and uninterrupted data collection process. Based on dynamic pattern recognition, intelligently trigger data collection tasks to improve the timeliness and accuracy of key data acquisition. Intelligently correct abnormal and missing values ​​based on patient history and clinical background, dynamically adjust standardized parameters to ensure data quality and consistency. Efficiently integrate multi-source numerical data, automatically detect and resolve data conflicts, and build a unified patient numerical view.

[0107] In summary, the method of the embodiment of the present application significantly improves the accuracy and real-time performance of clinical data analysis by introducing a dynamic weighting mechanism, an optimization algorithm, and a consistency assurance mechanism, and has important application value in precision medicine and personalized treatment.

[0108] It should be noted that the execution subject of the method for multi-center data processing in clinical trials provided in the embodiments of the present application may be a multi-center data processing device in clinical trials, or a control module in the multi-center data processing device in clinical trials for executing the method for multi-center data processing in clinical trials. In the embodiments of the present application, the method for multi-center data processing in clinical trials performed by a multi-center data processing device in clinical trials is taken as an example to illustrate the device for multi-center data processing in clinical trials provided in the embodiments of the present application.

[0109] like Figure 2 As shown, in a second aspect of an embodiment of the present application, a multi-center data processing device in a clinical trial is provided, and the device may include: An acquisition module 210, for acquiring multi-source clinical data of clinical trials; A first determination module 220 is used to determine the data type of each data in the multi-source clinical data, and dynamically adjust the weight of each data source according to the characteristics of the data type to obtain a first weight; The second determination module 230 is used to determine the clinical characteristics of individual patients in the multi-source clinical data, and dynamically adjust the weighting rules according to the clinical characteristics to obtain a first weighting rule; The processing module 240 is used to process the multi-source clinical data according to the first weight and the first weighting rule.

[0110] The multi-center data processing device in the clinical trial in the embodiment of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0111] The multi-center data processing device in the clinical trial in the embodiment of the present application may be a device with an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0112] The multi-center data processing device in the clinical trial provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented by the method embodiment are not described here.

[0113] Alternatively, if Figure 3 As shown, an embodiment of the present application also provides an electronic device 300, including a first processor 301, a first memory 302, and a program or instruction stored in the first memory 302 and executable on the first processor 301. When the program or instruction is executed by the first processor 301, each process of the embodiment of the multi-center data processing method in the above-mentioned clinical trial is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0114] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0115] Figure 4 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application.

[0116] The hardware structure 400 of the electronic device includes but is not limited to: a radio frequency unit 401, a network module 402, an audio output unit 403, an input unit 404, a sensor 405, a display unit 406, a user input unit 407, an interface unit 408, a second memory 409, and a second processor 410 and other components.

[0117] Those skilled in the art will appreciate that the hardware structure 400 of the electronic device may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the second processor 410 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Figure 4 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0118] It should be understood that in the embodiment of the present application, the input unit 404 may include a graphics processor (GPU) 4041 and a microphone 4042, and the graphics processor 4041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode. The display unit 406 may include a display panel 4061, and the display panel 4061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 407 includes a touch panel 4071 and other input devices 4072. The touch panel 4071 is also called a touch screen. The touch panel 4071 may include two parts: a touch detection device and a touch controller. Other input devices 4072 may include but are not limited to a physical keyboard, a function key (such as a volume control button, a switch button, etc.), a trackball, a mouse, and a joystick, which will not be repeated here. The second memory 409 can be used to store software programs and various data, including but not limited to applications and operating systems. The second processor 410 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, the user interface, and the application program, etc., and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the second processor 410 .

[0119] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the embodiment of the multi-center data processing method in the above-mentioned clinical trial is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0120] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0121] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the embodiment of the multi-center data processing method in the above-mentioned clinical trial, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0122] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0123] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0124] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0125] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for processing multi-center data in clinical trials, characterized in that: include: Access multi-source clinical data for clinical trials; Determine the data type of each data in the multi-source clinical data, and dynamically adjust the weight of each data source according to the characteristics of the data type to obtain a first weight; Determining clinical characteristics of individual patients in the multi-source clinical data, and dynamically adjusting weighting rules according to the clinical characteristics to obtain weighting rules; The multi-source clinical data are processed according to the first weights and the weighting rules.

2. The multi-center data processing method in clinical trials according to claim 1, characterized in that: After processing the multi-source clinical data according to the first weight and the weighting rule, the method for processing multi-center data in clinical trials further includes: Using machine learning and adaptive optimization algorithms to optimize the first weight to obtain a second weight; The multi-source clinical data is processed according to the second weight and the weighting rule.

3. The multi-center data processing method in clinical trials according to claim 2, characterized in that: The step of optimizing the first weight by using a machine learning and adaptive optimization algorithm to obtain a second weight includes: A multi-dimensional prediction model was constructed based on historical clinical data, and random forests were used to analyze the contribution of different features to clinical outcomes; The prediction model is trained using the multi-source clinical data processed according to the first weights and the weighting rules, so that the prediction model improves prediction accuracy.

4. The multi-center data processing method in clinical trials according to claim 3, characterized in that: The step of training the prediction model using the multi-source clinical data processed according to the first weight and the weighting rule includes: The entropy weight method and Bayesian optimization method are used to optimize the weight coefficients of each data source in the multi-source clinical data.

5. The multi-center data processing method in clinical trials according to claim 2, characterized in that: Before processing the multi-source clinical data according to the second weight and the weighting rule, the method for processing multi-center data in clinical trials further includes: Performing consistency analysis on the multi-source clinical data to obtain analysis results; The multi-source clinical data is adjusted for consistency according to the analysis results to obtain adjusted data.

6. The multi-center data processing method in clinical trials according to claim 5, characterized in that: After the consistency adjustment of the multi-source clinical data according to the analysis results, the multi-center data processing method in the clinical trial further includes: The adjusted data were verified for consistency using cross-validation and independent validation set methods to ensure data consistency.

7. The method for processing multi-center data in clinical trials according to claim 5, characterized in that: After processing the multi-source clinical data according to the second weight and the weighting rule, the method for processing multi-center data in clinical trials further includes: Obtain the latest clinical data of patients; The second weight, the weighting rule and the analysis result are adjusted according to the latest clinical data.

8. A multi-center data processing device in clinical trials, characterized in that: include: The acquisition module is used to obtain multi-source clinical data for clinical trials; A first determination module is used to determine the data type of each data in the multi-source clinical data, and dynamically adjust the weight of each data source according to the characteristics of the data type to obtain a first weight; A second determination module is used to determine clinical characteristics of individual patients in the multi-source clinical data, and dynamically adjust the weighting rule according to the clinical characteristics to obtain a first weighting rule; A processing module is used to process the multi-source clinical data according to the first weight and the first weighting rule.

9. An electronic device, characterized in that: include: A processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the multi-center data processing method in a clinical trial as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores programs or instructions, and when the programs or instructions are executed by the processor, the multi-center data processing method in a clinical trial as described in any one of claims 1-7 is implemented.

Images