System and method for detecting potency of pancreatic elastase 1 test reagent, medium

Abstract

The present application discloses a pancreatic elastase 1 detection reagent performance detection system and method, medium, relates to the field of biomedical detection related, including: reagent monitoring module is used for real-time monitoring and acquiring pancreatic elastase 1 detection reagent performance correlation data;Data mining module is used for extracting key factors to performance correlation data, and mining and acquiring reagent performance factor dataset;Channel building module is used for training fusion optimization to reagent performance factor dataset using time series network structure, builds reagent performance prediction channel, and carries out prediction analysis to performance correlation data;Dynamic optimization module is used for constructing performance optimization strategy space according to target reagent performance information, and dynamically optimizing storage environment and use detection parameters.Solve the technical problems that the performance detection of existing pancreatic elastase 1 detection reagent cannot be monitored in real time and the detection performance cannot be comprehensively optimized, and achieve the technical effect of ensuring the stability of reagent detection performance.

Description

Technical Field

[0001] This application relates to the field of biomedical testing, and in particular to a system, method, and medium for detecting the efficacy of a pancreatic elastase 1 detection reagent. Background Technology

[0002] The stability of the efficacy of pancreatic elastase 1 (PEP1) assay reagents is crucial for ensuring accurate and reliable test results. Currently, the main methods for addressing the issue of stable efficacy in PEP1 assay reagents are periodic quality testing of the reagents and setting and adjusting storage conditions and usage parameters based on routine experience. However, current methods, due to the time intervals between periodic testing, cannot monitor changes in reagent efficacy in real time, and routine experience is insufficient to comprehensively cover all dynamic factors affecting reagent efficacy, making it difficult to maintain consistently stable assay efficacy.

[0003] Currently, the efficacy of pancreatic elastase 1 detection reagents faces technical challenges, including the inability to monitor and fully optimize detection performance in real time. Summary of the Invention

[0004] This application provides a system, method, and medium for detecting the efficacy of pancreatic elastase 1 (PA1) test reagents. It employs methods such as acquiring PA1 test reagent efficacy-related data, extracting key factors of reagent efficacy from the data to form a key factor set, mining and obtaining a dataset of reagent efficacy factors, using a time-series network structure to train, fuse, and optimize the dataset, building a reagent efficacy prediction channel, predictively analyzing efficacy-related data based on the prediction channel, outputting target reagent efficacy information, constructing an efficacy optimization strategy space based on the efficacy information, and dynamically optimizing the reagent storage environment and usage detection parameters. These techniques solve the technical problems of existing PA1 test reagent efficacy detection methods, which cannot monitor and comprehensively optimize detection efficacy in real time, thus achieving the technical effect of ensuring stable reagent detection efficacy.

[0005] This application provides a efficacy detection system for a pancreatic elastase 1 (PA1) detection reagent, comprising: a reagent monitoring module for real-time monitoring and acquisition of efficacy-related data of the PA1 detection reagent, wherein the efficacy-related data includes environmental parameters, production batch number, and instrument parameters; a data mining module for extracting key factors from the efficacy-related data to obtain a reagent efficacy key factor set, and mining a reagent efficacy factor dataset based on the reagent efficacy key factor set; a channel construction module for training, fusing, and optimizing the reagent efficacy factor dataset using a time series network structure to build a reagent efficacy prediction channel, performing predictive analysis on the efficacy-related data based on the reagent efficacy prediction channel, and outputting target reagent efficacy information; and a dynamic optimization module for constructing an efficacy optimization strategy space based on the target reagent efficacy information, and dynamically optimizing the storage environment and detection parameters of the PA1 detection reagent based on the efficacy optimization strategy space.

[0006] In a possible implementation, the data mining module includes: a data preprocessing module, used to perform abnormal data identification and cleaning filtering on the performance-related data according to data application standards to obtain standard performance-related data; an influencing factor extraction module, used to extract influencing factors from the standard performance-related data to obtain a set of related influencing factors and a corresponding reagent performance dataset; a correlation coefficient calculation module, used to calculate the correlation coefficient set of each influencing factor in the set of related influencing factors and the influencing factor correlation coefficient set of the reagent performance dataset; and a key factor screening module, used to screen key factors in the set of related influencing factors based on the correlation coefficient set of influencing factors to obtain the reagent performance key factor set.

[0007] In a possible implementation, the data mining module includes: a correlation feature acquisition module, used to obtain a correlation feature set of efficacy factors based on the set of key factors of reagent efficacy; a feature tree construction module, used to perform cascade analysis and feature tree construction on the correlation feature set of efficacy factors respectively to obtain a set of cascaded feature trees of efficacy factors; and a correlation data mining module, used to perform correlation data mining based on the set of cascaded feature trees of efficacy factors to obtain a dataset of reagent efficacy factors.

[0008] In a possible implementation, the association data mining module includes: a data volume requirement hierarchy determination module, used to determine the data volume requirement hierarchy of efficacy factors based on the reagent efficacy prediction target; an association depth analysis module, used to perform association depth analysis on the cascaded feature tree set of efficacy factors according to the data volume requirement hierarchy of efficacy factors to obtain the feature tree depth of efficacy factors; and a feature data mining module, used to perform feature data mining based on the feature tree depth of efficacy factors to obtain the reagent efficacy factor dataset.

[0009] In a possible implementation, the channel construction module includes: a prediction training module, used to perform prediction training on the reagent efficacy factor dataset using a time series network structure to generate an efficacy factor prediction branch network set; a prediction performance verification module, used to verify the prediction performance of the efficacy factor prediction branch network set and determine the branch network weight factor information based on the branch network performance verification results; and a weighted fusion module, used to perform weighted fusion and test optimization on the efficacy factor prediction branch network set based on the branch network weight factor information to construct a reagent efficacy prediction channel.

[0010] In a possible implementation, the weighted fusion module includes: an initial efficacy prediction channel acquisition module, used to perform weighted fusion of the efficacy factor prediction branch network set based on the branch network weight factor information to obtain an initial efficacy prediction channel; a test analysis module, used to perform test analysis on the initial efficacy prediction channel using an efficacy test dataset to obtain channel prediction effect parameters; and an iterative optimization module, used to select a channel optimizer based on the channel prediction effect parameters, and iteratively optimize the initial efficacy prediction channel based on the channel optimizer to construct the reagent efficacy prediction channel.

[0011] In a possible implementation, the dynamic optimization module includes: a performance optimization strategy rule base construction module, used to construct a performance optimization strategy rule base, the performance optimization strategy rule base including performance problem triggering conditions and corresponding performance optimization strategies, wherein the performance optimization strategies are specifically optimization strategies for reagent storage environment and detection parameters; an optimization strategy matching module, used to perform performance problem triggering and optimization strategy matching on the performance optimization strategy rule base based on the target reagent performance information to obtain a target performance optimization strategy; and an optimization threshold parsing module, used to perform optimization threshold parsing on the target reagent performance information based on the target performance optimization strategy to construct the performance optimization strategy space.

[0012] In a possible implementation, the dynamic optimization module includes: a global optimization module, used to perform global optimization on the performance optimization strategy space according to the performance optimization effect evaluation index set, and determine the target performance optimization strategy parameters; and a reagent dynamic strategy control module, used to perform strategy optimization feedback on the storage environment and usage detection parameters based on the target performance optimization strategy parameters, obtain reagent performance feedback parameters, and use a PID controller to perform reagent dynamic strategy control based on the reagent performance feedback parameters.

[0013] This application also provides a method for detecting the efficacy of a pancreatic elastase 1 (PEP1) detection reagent, comprising: real-time monitoring and acquisition of efficacy-related data of the PEP1 detection reagent, wherein the efficacy-related data includes environmental parameters, production batch number, and instrument parameters; extraction of key factors from the efficacy-related data to obtain a set of key factors for reagent efficacy, and mining a dataset of reagent efficacy factors based on the set of key factors for reagent efficacy; training, fusion, and optimization of the dataset of reagent efficacy factors using a time-series network structure to build a reagent efficacy prediction channel; performing predictive analysis on the efficacy-related data based on the reagent efficacy prediction channel to output target reagent efficacy information; constructing an efficacy optimization strategy space based on the target reagent efficacy information; and dynamically optimizing the storage environment and detection parameters of the PEP1 detection reagent based on the efficacy optimization strategy space.

[0014] This application also provides a computer-readable storage medium, including: a computer program stored thereon, which, when executed by a processor, implements a method for detecting the efficacy of a pancreatic elastase 1 assay reagent.

[0015] This application proposes a system, method, and medium for detecting the efficacy of a pancreatic elastase 1 (PEP1) test reagent. A reagent monitoring module acquires real-time efficacy-related data of the PEP1 test reagent, including environmental parameters, production batch number, and instrument parameters. A data mining module extracts key factors from the efficacy-related data to obtain a set of key factors for reagent efficacy. Based on this set, a dataset of reagent efficacy factors is mined. A channel building module uses a time-series network structure to train, fuse, and optimize the dataset, building a reagent efficacy prediction channel. Based on this prediction channel, the efficacy-related data is predicted and analyzed to output target reagent efficacy information. A dynamic optimization module constructs an efficacy optimization strategy space based on the target reagent efficacy information. Based on this strategy space, the storage environment and usage parameters of the PEP1 test reagent are dynamically optimized. This achieves the technical effect of ensuring stable reagent detection efficacy. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1This is a schematic diagram of the structure of the efficacy detection system for the pancreatic elastase 1 detection reagent provided in the embodiments of this application.

[0018] Figure 2 This is a schematic flowchart illustrating the efficacy detection method of the pancreatic elastase 1 detection reagent provided in the embodiments of this application.

[0019] Figure labeling: 10 reagent monitoring module, 20 data mining module, 30 channel construction module, 40 dynamic optimization module. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments; however, it is understood that "some embodiments" may be the same or different subsets of all possible embodiments and may be combined with each other without conflict. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides a performance detection system for a pancreatic elastase 1 detection reagent, such as... Figure 1 As shown, the system includes:

[0024] The reagent monitoring module 10 is used to monitor and acquire the efficacy correlation data of the pancreatic elastase 1 detection reagent in real time, wherein the efficacy correlation data includes environmental parameters, production batch number and instrument parameters.

[0025] Specifically, the performance-related data refers to various data related to the performance of the pancreatic elastase 1 test kit, including environmental parameters (temperature, humidity, light intensity, etc.), production batch number, and instrument parameters (detection wavelength, sensitivity, calibration status, etc.). Specifically, temperature and humidity sensors, light sensors, etc., are used to collect parameters such as temperature, humidity, and light intensity of the storage environment of the pancreatic elastase 1 test kit in real time. The production batch number information of the pancreatic elastase 1 test kit is quickly and accurately obtained through a barcode scanner or RFID reader. Communication with the control system of the testing instrument is established, and the instrument's operating parameters, such as detection wavelength, sensitivity, and calibration status, are read in real time through data interfaces (such as USB, RS232, network interface, etc.).

[0026] For example, in environmental parameter monitoring, a DHT11 temperature and humidity sensor is used to collect temperature and humidity data of the storage environment of the pancreatic elastase 1 test reagent every 5 minutes, and the data is transmitted to the data acquisition system via I2C bus. For production batch number identification, a Honeywell 1900 series barcode scanner is used to scan the barcode on the pancreatic elastase 1 test reagent packaging to obtain the production batch number information, which is then stored in the database. Regarding instrument parameter acquisition, the instrument's detection wavelength settings, sensitivity level, and other parameters are read in real time via its API interface after connecting to the testing instrument's USB interface.

[0027] The data mining module 20 is used to extract key factors from the efficacy-related data to obtain a set of key factors for reagent efficacy, and to mine and obtain a dataset of reagent efficacy factors based on the set of key factors for reagent efficacy.

[0028] Specifically, key factor extraction refers to screening out the key factors that significantly impact reagent efficacy from a large amount of efficacy-related data for model training and prediction. Machine learning algorithms, such as Principal Component Analysis (PCA) and Random Forest, are used to extract features from the collected efficacy-related data, screening out the key factors that significantly affect reagent efficacy. For example, PCA is used to reduce the dimensionality of collected environmental parameters (temperature, humidity, light intensity), production batch numbers, and instrument parameters (detection wavelength, sensitivity, etc.), extracting the main components as key factors. Based on the extracted key factors, a reagent efficacy factor dataset is constructed through correlation factor mining, providing a data foundation for predictive model training. When constructing the reagent efficacy factor dataset, the extracted factors are correlated with the actual efficacy data (such as detection accuracy, stability, etc.) of the pancreatic elastase 1 detection reagent, forming a dataset containing multiple features and labels for model training.

[0029] In one possible implementation, the data mining module 20 includes: a data preprocessing module, used to perform abnormal data identification and cleaning filtering on the performance-related data according to data application standards to obtain standard performance-related data; an influencing factor extraction module, used to extract influencing factors from the standard performance-related data to obtain a set of related influencing factors and a corresponding reagent performance dataset; a correlation coefficient calculation module, used to calculate the correlation coefficient set of each influencing factor in the set of related influencing factors and the influencing factor correlation coefficient set of the reagent performance dataset; and a key factor screening module, used to screen key factors in the set of related influencing factors based on the correlation coefficient set of influencing factors to obtain the reagent performance key factor set.

[0030] Specifically, statistical methods (such as Z-score and IQR) or machine learning methods (such as Isolation Forest) are used to identify outlier data. The identified outlier data is then processed, such as by deleting, correcting, or imputing missing values. For example, using the Z-score method to identify outlier data, for each environmental parameter (such as temperature and humidity), its mean and standard deviation are calculated, and then the Z-score for each data point is calculated. If the Z-score is greater than 3 or less than -3, the data point is considered an outlier. For identified outliers, median imputation is used. If a temperature data point is identified as an outlier, it is replaced with the median of other normal temperature data within the same time period.

[0031] Principal component analysis (PCA) or linear regression are used to extract influencing factors from standard efficacy correlation data. These extracted factors are then correlated with reagent efficacy data to form a dataset. For example, PCA can be used to reduce the dimensionality of standard efficacy correlation data, extracting the main components as influencing factors. These extracted factors (such as temperature, humidity, and production batch number) are then correlated with the actual efficacy data of the reagents (such as detection accuracy and stability) to form a dataset containing multiple features and labels.

[0032] The correlation coefficients between each influencing factor and the reagent efficacy dataset are calculated using methods such as Pearson correlation coefficient or Spearman rank correlation coefficient. For example, the correlation coefficient between temperature and detection accuracy is calculated using Pearson correlation coefficient. Assuming the correlation coefficient between temperature and detection accuracy is 0.85, it indicates a strong positive correlation between the two.

[0033] Based on the set of correlation coefficients of influencing factors, factors with correlation coefficients higher than a certain threshold (e.g., 0.8) are selected as key factors, and these key factors are combined into a set of key factors for reagent efficacy. This approach, through data preprocessing, influencing factor extraction, correlation coefficient calculation, and key factor selection, effectively extracts key factors from a large amount of efficacy-related data, providing a high-quality data foundation for training the reagent efficacy prediction model and improving the model's predictive performance and interpretability.

[0034] In one possible implementation, the data mining module 20 includes: an association feature acquisition module, used to obtain an association feature set of efficacy factors based on the set of key factors of reagent efficacy; a feature tree construction module, used to perform cascade analysis and feature tree construction on the association feature set of efficacy factors respectively to obtain a set of cascaded feature trees of efficacy factors; and an association data mining module, used to perform association data mining based on the set of cascaded feature trees of efficacy factors to obtain a dataset of reagent efficacy factors.

[0035] Specifically, based on the set of key factors for reagent efficacy, correlation features between efficacy factors are extracted through statistical analysis or machine learning methods. For example, assuming the set of key factors for reagent efficacy includes temperature, humidity, and instrument parameters, correlation features are extracted by analyzing the interrelationships between these factors, such as the interaction between temperature and humidity, and the interaction between temperature and instrument parameters. Using a multinomial feature expansion method, the interaction terms of temperature, humidity, and instrument parameters are used as correlation features.

[0036] Cascade analysis was performed on the extracted associated features to analyze the hierarchical relationships between different features. Based on the cascade analysis results, a cascade feature tree of efficacy factors was constructed, forming a set of cascade feature trees of efficacy factors. For example, cascade analysis revealed that the interaction between temperature and humidity has a greater impact on reagent efficacy than the interaction between temperature and instrument parameters. Based on this hierarchical relationship, the constructed feature tree is as follows: Root node: Temperature; First level: Humidity (interaction with temperature); Second level: Instrument parameters (interaction with temperature). Table 1 shows an example of the constructed feature tree.

[0037] Table 1: Example of a feature tree

[0038] Feature tree hierarchy Key factors Association features Parent node root node temperature - - First level humidity Temperature × Humidity temperature Second level Instrument parameters Temperature × Instrument Parameters temperature

[0039] Based on a cascaded feature tree set of efficacy factors, data mining algorithms (such as Apriori and FP-Growth) are used to mine associated data to obtain a reagent efficacy factor dataset. For example, the Apriori algorithm is used to mine association rules, identifying frequent itemsets and strong association rules. Suppose the mined rules include: "If the temperature is around 25℃ and the humidity is around 60%, the reagent detection accuracy is high"; "If the temperature is around 25℃ and the instrument parameter = 400nm, the reagent detection stability is high"; "If the humidity is around 60% and the instrument parameter = 400nm, the detection accuracy is high". The mined association rules are organized into a reagent efficacy factor dataset for model training and prediction. This approach effectively extracts association features from the set of key reagent efficacy factors, constructs feature trees, and mines valuable association rules through association feature acquisition, feature tree construction, and association data mining. This not only enriches the data's content but also improves the model's interpretability and predictive performance, providing strong support for optimizing reagent efficacy.

[0040] In one possible implementation, the association data mining module includes: a data volume requirement hierarchy determination module, used to determine the data volume requirement hierarchy of efficacy factors based on the reagent efficacy prediction target; an association depth analysis module, used to perform association depth analysis on the cascaded feature tree set of efficacy factors according to the data volume requirement hierarchy of efficacy factors to obtain the feature tree depth of efficacy factors; and a feature data mining module, used to perform feature data mining based on the feature tree depth of efficacy factors to obtain the reagent efficacy factor dataset.

[0041] Specifically, the required data volume level is determined based on the reagent efficacy prediction target. For example, a higher accuracy prediction target requires more data to support model training. Data volume requirements are divided into different levels, such as low, medium, and high, each corresponding to different data volume requirements. For instance, for reagent efficacy prediction targets, if the target is high-accuracy prediction (e.g., error less than 5%), the data volume requirement level is high; if the target is medium-accuracy prediction (e.g., error less than 10%), the data volume requirement level is medium; and if the target is preliminary prediction (e.g., error less than 20%), the data volume requirement level is low. The low-level data volume requirement is set at 1000 data points, the medium-level at 5000 data points, and the high-level at 10000 data points.

[0042] Based on the hierarchical data volume requirements of performance factors, a deep analysis is performed on the cascaded feature tree set of performance factors. By analyzing the hierarchical structure of the feature trees, the depth of each feature tree is determined to ensure that the data volume meets the prediction target requirements. For example, for high-level data volume requirements (10,000 data points), the depth of the feature trees is analyzed to ensure that each feature tree can provide sufficient data points. Assuming the feature tree depth is 3 levels, the data volume requirements for each level are 3,000, 3,000, and 4,000 data points, respectively. For mid-level data volume requirements (5,000 data points), the feature tree depth can be 2 levels, with data volume requirements for each level being 2,000 and 3,000 data points, respectively.

[0043] Based on a determined feature tree depth for efficacy factors, data mining algorithms (such as Apriori and FP-Growth) are used to perform feature data mining to obtain a reagent efficacy factor dataset. The mined feature data is then organized into a dataset for model training and prediction. For example, for a high-level data requirement (10,000 data points), the Apriori algorithm is used to mine association rules with a feature tree depth of 3. These association rules are then organized into a reagent efficacy factor dataset for model training and prediction. This approach, by determining the data requirement level, analyzing association depth, and performing feature data mining, can rationally allocate data resources, ensuring that the data volume meets the prediction target requirements and improving the efficiency and accuracy of data mining.

[0044] The channel construction module 30 is used to train, fuse and optimize the reagent efficacy factor dataset using a time series network structure, build a reagent efficacy prediction channel, perform predictive analysis on the efficacy-related data based on the reagent efficacy prediction channel, and output target reagent efficacy information.

[0045] Specifically, time series network structures refer to neural network structures specifically designed for processing time series data, such as Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs), which can learn the time series characteristics and patterns in the data. Using time series neural network structures such as LSTM or GRUs, a reagent efficacy factor dataset is trained to learn the time series characteristics and patterns in the data. The model is trained, fused, and optimized through methods such as cross-validation and grid search to improve its predictive performance. The trained model is then deployed as a reagent efficacy prediction channel, receiving efficacy-related data in real time and outputting target reagent efficacy information.

[0046] For example, using an LSTM network architecture, the reagent efficacy factor dataset is divided according to time series and input into the LSTM network for training. During training, the Adam optimizer is used with a learning rate of 0.001 and 100 training epochs. The model is evaluated using cross-validation, dividing the dataset into training and validation sets, and using validation set performance metrics (such as mean squared error and accuracy) to select the optimal model parameters. The trained LSTM model is deployed as a reagent efficacy prediction channel, receiving real-time data such as temperature and humidity of the storage environment of the pancreatic elastase 1 detection reagent, as well as instrument parameters such as detection wavelength, sensitivity, reagent dosage, and incubation time, and outputting the reagent efficacy prediction results, such as the predicted value of detection accuracy.

[0047] In one possible implementation, the channel construction module 30 includes: a prediction training module, used to perform prediction training on the reagent efficacy factor dataset using a time series network structure to generate an efficacy factor prediction branch network set; a prediction performance verification module, used to verify the prediction performance of the efficacy factor prediction branch network set and determine the branch network weight factor information based on the branch network performance verification results; and a weighted fusion module, used to perform weighted fusion and test optimization on the efficacy factor prediction branch network set based on the branch network weight factor information to construct a reagent efficacy prediction channel.

[0048] Specifically, the reagent efficacy factor dataset is divided into multiple subsets, with each subset used to train a branch network. Different data preprocessing methods, feature selection methods, or network structures are applied to each subset to generate multiple different branch networks. Through these methods, multiple efficacy factor prediction branch networks are generated, forming a branch network set. For example, assuming the dataset contains 10,000 records, it can be divided into 5 subsets, each containing 2,000 records, and the generated branch networks are shown in Table 2.

[0049] Table 2: Examples of Branch Network Sets for Performance Factor Prediction

[0050]

[0051] Performance validation was performed on each performance factor prediction branch network, and the model performance was evaluated using metrics such as mean squared error (MSE) and accuracy. Based on the performance validation results, weight factors were assigned to each branch network. The weight factors are proportional to the model performance. An example of weight factor allocation is shown in Table 3.

[0052] Table 3: Example of Weight Factor Allocation

[0053] Branch network MSE accuracy Weighting factors Network 1 0.05 0.85 0.25 Network 2 0.06 0.83 0.20 Network 3 0.04 0.84 0.22 Network 4 0.03 0.86 0.28 Network 5 0.04 0.87 0.25

[0054] Based on the weight factor information of the branch networks, the efficacy factor prediction branch networks are weighted and fused to construct a comprehensive reagent efficacy prediction channel. The fused prediction channel is then tested and optimized to ensure its accuracy and stability in practical applications. For example, the formula for the fused prediction channel can be: Comprehensive efficacy prediction = ∑(Efficacy factor prediction branch network output × Weight factor). This approach, by training multiple branch networks using different partitioning methods and processing techniques, fully utilizes the information in the dataset, improving the model's diversity and robustness. Through performance validation and weight factor allocation, each branch network can be reasonably weighted according to model performance, improving the accuracy and reliability of the comprehensive prediction.

[0055] In one possible implementation, the weighted fusion module includes: an initial efficacy prediction channel acquisition module, used to perform weighted fusion of the efficacy factor prediction branch network set based on the branch network weight factor information to obtain an initial efficacy prediction channel; a test analysis module, used to perform test analysis on the initial efficacy prediction channel using an efficacy test dataset to obtain channel prediction effect parameters; and an iterative optimization module, used to select a channel optimizer based on the channel prediction effect parameters, and iteratively optimize the initial efficacy prediction channel based on the channel optimizer to construct the reagent efficacy prediction channel.

[0056] Specifically, based on the branch network weight factor information, the performance factor prediction branch network set is weighted and fused to generate initial performance prediction channels. The initial performance prediction channels are tested and analyzed using an independent performance test dataset, and channel prediction performance parameters, such as mean squared error (MSE), accuracy, and recall, are calculated. Based on the channel prediction performance parameters, a suitable channel optimizer, such as Adam or SGD, is selected. The selected optimizer is used to iteratively fine-tune the initial performance prediction channels, with the optimization goal of reducing MSE and improving accuracy. Through multiple iterations, the model parameters are gradually adjusted until the preset performance indicators are reached or convergence occurs. For example, the Adam optimizer is selected with a learning rate of 0.001 and 100 iterations. In each iteration, the loss function on the training set is calculated, and the model parameters are updated. This implementation method, through testing, analysis, and iterative tuning, gradually optimizes the performance of the prediction channels, improving the accuracy and stability of predictions.

[0057] The dynamic optimization module 40 is used to construct a performance optimization strategy space based on the target reagent performance information, and to dynamically optimize the storage environment and detection parameters of the pancreatic elastase 1 detection reagent based on the performance optimization strategy space.

[0058] Specifically, based on the target reagent efficacy information and considering the adjustable range of the storage environment and detection parameters used, an efficacy optimization strategy space is constructed. This space encompasses all possible adjustment strategies for the storage environment and detection parameters. Reinforcement learning algorithms, such as Q-learning and deep reinforcement learning, are employed to dynamically adjust the storage environment and detection parameters based on the current reagent efficacy prediction results, thereby optimizing reagent efficacy.

[0059] For example, when constructing the performance optimization strategy space, the temperature range for reagent storage is set to 2℃–8℃, the humidity range to 30%–70%, the instrument detection wavelength range to 400nm–700nm, the sensitivity range to low, medium, and high, the reagent dosage range to 0.5mL–2.0mL, and the incubation time range to 15 minutes–120 minutes. Using the Q-learning algorithm, the reagent performance prediction results are used as environmental feedback. Based on the current storage environment parameters and instrument parameters, the optimal adjustment strategy is dynamically selected. For example, when the prediction results indicate low reagent detection accuracy, the reagent performance can be optimized by adjusting the storage environment temperature and humidity, as well as the instrument detection wavelength, sensitivity, reagent dosage, and incubation time.

[0060] In one possible implementation, the dynamic optimization module 40 includes: an efficiency optimization strategy rule base construction module, used to construct an efficiency optimization strategy rule base, the efficiency optimization strategy rule base including efficiency problem triggering conditions and corresponding efficiency optimization strategies, wherein the efficiency optimization strategies are specifically optimization strategies for reagent storage environment and use detection parameters; an optimization strategy matching module, used to perform efficiency problem triggering and optimization strategy matching on the efficiency optimization strategy rule base based on the target reagent efficiency information to obtain a target efficiency optimization strategy; and an optimization threshold parsing module, used to perform optimization threshold parsing on the target reagent efficiency information based on the target efficiency optimization strategy to construct the efficiency optimization strategy space.

[0061] Specifically, a performance optimization strategy rule base is constructed, which includes performance problem triggering conditions and corresponding performance optimization strategies. The optimization strategies are specifically for optimizing reagent storage environment (such as temperature and humidity) and detection parameters (such as wavelength, sensitivity, reagent dosage, and incubation time). Examples of performance optimization strategy rule bases are shown in Table 4.

[0062] Table 4: Examples of Performance Optimization Strategy Rule Base

[0063]

[0064] Based on the target reagent performance information, the corresponding performance issue is triggered. The appropriate optimization strategy is then matched against the performance optimization strategy rule base to obtain the target performance optimization strategy. For example, assuming the current target reagent performance information shows a detection accuracy of 78%, rule 1 is triggered according to the performance optimization strategy rule base. The matched target performance optimization strategy is: temperature = 4℃, humidity = 50%, wavelength = 500nm, sensitivity = high, reagent volume = 1.5mL, incubation time = 60 minutes.

[0065] Based on the target performance optimization strategy, the target reagent performance information is analyzed using optimization thresholds to determine the adjustable range of each parameter, forming a complete performance optimization strategy space. For example, suppose the target performance optimization strategy is: temperature = 4℃, humidity = 50%, wavelength = 500nm, sensitivity = high, reagent volume = 1.5mL, incubation time = 60 minutes. The optimization threshold analysis module determines the adjustable range of these parameters as follows: temperature: 3℃~5℃; humidity: 45%~55%; wavelength: 480nm~520nm; sensitivity: high; reagent volume: 1.4mL~1.6mL; incubation time: 50 minutes~70 minutes. This implementation method, by constructing a performance optimization strategy rule base, quickly matches optimization strategies and constructs a complete performance optimization strategy space through optimization threshold analysis, providing strong support for the dynamic optimization of reagent performance and ensuring the operability and flexibility of the optimization process.

[0066] In one possible implementation, the dynamic optimization module 40 includes: a global optimization module, used to perform global optimization on the performance optimization strategy space according to the performance optimization effect evaluation index set, and determine the target performance optimization strategy parameters; and a reagent dynamic strategy control module, used to perform strategy optimization feedback on the storage environment and usage detection parameters based on the target performance optimization strategy parameters, obtain reagent performance feedback parameters, and use a PID controller to perform reagent dynamic strategy control based on the reagent performance feedback parameters.

[0067] Specifically, optimization algorithms (such as genetic algorithms and particle swarm optimization) are used to globally optimize the performance optimization strategy space. A set of evaluation metrics (such as detection accuracy, stability, and response time) is used to assess the effectiveness of the optimization strategy and determine the target performance optimization strategy parameters. For example, suppose the performance optimization strategy space includes the adjustable range of parameters such as temperature, humidity, wavelength, sensitivity, reagent dosage, and incubation time. A genetic algorithm is used for global optimization, and the evaluation metrics include detection accuracy, stability, and response time. Through multiple iterations, the optimal parameter combination is found.

[0068] Based on the target performance optimization strategy parameters determined by the global optimization module, the storage environment and usage detection parameters are optimized and feedback is provided to obtain reagent performance feedback parameters. A PID controller is used for dynamic strategy control based on the reagent performance feedback parameters to ensure the stability and accuracy of parameter adjustment. For example, suppose the optimal parameter combination determined by the global optimization module is: temperature = 4.1℃, humidity = 51%, wavelength = 505nm, sensitivity = high, reagent volume = 1.55mL, incubation time = 62 minutes. The PID controller is used to dynamically control these parameters to ensure the stability and accuracy of parameter adjustment. Through a real-time feedback mechanism, the parameters are adjusted to achieve optimal performance. This implementation method, through global optimization and dynamic control by a PID controller, ensures the optimization of reagent performance and the stability of the system.

[0069] This application embodiment acquires efficacy-related data of pancreatic elastase 1 detection reagents, extracts key factors of reagent efficacy from the data to form a key factor set, and mines and acquires a dataset of reagent efficacy factors. It then uses a time-series network structure to train, fuse, and optimize the dataset, constructing a reagent efficacy prediction channel. Based on the prediction channel, it performs predictive analysis on efficacy-related data, outputs target reagent efficacy information, and constructs an efficacy optimization strategy space based on the efficacy information. This involves dynamically optimizing the reagent storage environment and usage detection parameters, thus solving the technical problem of existing pancreatic elastase 1 detection reagents' inability to monitor and comprehensively optimize detection efficacy in real time, achieving the technical effect of ensuring stable reagent detection efficacy.

[0070] In the above text, refer to Figure 1 A performance detection system for a pancreatic elastase 1 detection reagent according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 A method for detecting the efficacy of a pancreatic elastase 1 detection reagent according to an embodiment of the present invention is described.

[0071] The efficacy detection method for the pancreatic elastase 1 detection reagent according to embodiments of the present invention is used to solve the technical problems of existing pancreatic elastase 1 detection reagents in terms of inability to monitor and comprehensively optimize detection efficacy in real time, thereby achieving the technical effect of ensuring stable detection efficacy of the reagent.

[0072] The efficacy detection method for a pancreatic elastase 1 (PA1) assay reagent includes: real-time monitoring and acquisition of efficacy-related data for the PA1 assay reagent, wherein the efficacy-related data includes environmental parameters, production batch number, and instrument parameters; extraction of key factors from the efficacy-related data to obtain a set of key factors for reagent efficacy, and mining a dataset of reagent efficacy factors based on the set of key factors for reagent efficacy; training, fusion, and optimization of the dataset of reagent efficacy factors using a time-series network structure to build a reagent efficacy prediction channel; predictive analysis of the efficacy-related data based on the reagent efficacy prediction channel to output target reagent efficacy information; construction of an efficacy optimization strategy space based on the target reagent efficacy information; and dynamic strategy optimization of the storage environment and detection parameters of the PA1 assay reagent based on the efficacy optimization strategy space.

[0073] The process of obtaining the key factor set for reagent efficacy may further include: identifying and cleaning / filtering abnormal data in the efficacy-related data according to data application standards to obtain standard efficacy-related data; extracting influencing factors from the standard efficacy-related data to obtain a set of related influencing factors and a corresponding reagent efficacy dataset; calculating the correlation coefficient set of each influencing factor in the set of related influencing factors and the influencing factor set of the reagent efficacy dataset; and screening key factors in the set of related influencing factors based on the correlation coefficient set of influencing factors to obtain the key factor set for reagent efficacy.

[0074] The step of mining the reagent efficacy factor dataset based on the set of key factors of reagent efficacy may further include: obtaining an efficacy factor association feature set based on the set of key factors of reagent efficacy; performing cascade analysis and feature tree construction on the efficacy factor association feature set to obtain an efficacy factor cascade feature tree set; and performing association data mining based on the efficacy factor cascade feature tree set to obtain the reagent efficacy factor dataset.

[0075] The process of obtaining the reagent efficacy factor dataset may further include: determining the required data volume hierarchy of efficacy factors based on the reagent efficacy prediction target; performing a correlation depth analysis on the cascaded feature tree set of efficacy factors according to the required data volume hierarchy of efficacy factors to obtain the feature tree depth of efficacy factors; and performing feature data mining based on the feature tree depth of efficacy factors to obtain the reagent efficacy factor dataset.

[0076] The process of building a reagent efficacy prediction channel may further include: using a time series network structure to perform prediction training on the reagent efficacy factor dataset to generate an efficacy factor prediction branch network set; verifying the prediction performance of the efficacy factor prediction branch network set and determining the branch network weight factor information based on the branch network performance verification results; and performing weighted fusion and test optimization on the efficacy factor prediction branch network set based on the branch network weight factor information to construct a reagent efficacy prediction channel.

[0077] The construction of the reagent efficacy prediction channel may further include: weighting and fusing the efficacy factor prediction branch network set based on the branch network weight factor information to obtain an initial efficacy prediction channel; testing and analyzing the initial efficacy prediction channel using an efficacy test dataset to obtain channel prediction effect parameters; selecting a channel optimizer based on the channel prediction effect parameters, and iteratively optimizing the initial efficacy prediction channel based on the channel optimizer to construct the reagent efficacy prediction channel.

[0078] The step of constructing an efficiency optimization strategy space based on the target reagent efficiency information may further include: constructing an efficiency optimization strategy rule base, which includes efficiency problem triggering conditions and corresponding efficiency optimization strategies, wherein the efficiency optimization strategies are specifically optimization strategies for reagent storage environment and usage detection parameters; matching efficiency problem triggering and optimization strategies based on the target reagent efficiency information to obtain a target efficiency optimization strategy; and performing optimization threshold analysis on the target reagent efficiency information based on the target efficiency optimization strategy to construct the efficiency optimization strategy space.

[0079] The step of dynamically optimizing the storage environment and usage detection parameters of the pancreatic elastase 1 detection reagent based on the performance optimization strategy space may further include: performing global optimization on the performance optimization strategy space according to the performance optimization effect evaluation index set to determine the target performance optimization strategy parameters; performing strategy optimization feedback on the storage environment and usage detection parameters based on the target performance optimization strategy parameters to obtain reagent performance feedback parameters; and using a PID controller to perform dynamic strategy control of the reagent based on the reagent performance feedback parameters.

[0080] The efficacy detection system for the pancreatic elastase 1 detection reagent provided in this embodiment of the invention can execute the efficacy detection method for the pancreatic elastase 1 detection reagent provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0081] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0082] Based on the foregoing embodiments, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor of an electronic device, can implement the efficacy detection method of the pancreatic elastase 1 detection reagent as described in any of the preceding embodiments.

[0083] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A performance detection system for a pancreatic elastase 1 assay reagent, characterized in that, The system includes: The reagent monitoring module is used to monitor and acquire the efficacy correlation data of the pancreatic elastase 1 detection reagent in real time, wherein the efficacy correlation data includes environmental parameters, production batch number and instrument parameters; The data mining module is used to extract key factors from the performance-related data to obtain a set of key factors for reagent performance, and to mine and obtain a dataset of reagent performance factors based on the set of key factors for reagent performance. The channel building module is used to train, fuse, and optimize the reagent efficacy factor dataset using a time series network structure, build a reagent efficacy prediction channel, perform predictive analysis on the efficacy-related data based on the reagent efficacy prediction channel, and output target reagent efficacy information. The dynamic optimization module is used to construct a performance optimization strategy space based on the target reagent performance information, and to dynamically optimize the storage environment and detection parameters of the pancreatic elastase 1 detection reagent based on the performance optimization strategy space. The channel construction module includes: The prediction training module is used to perform prediction training on the reagent efficacy factor dataset using a time series network structure, and generate a set of efficacy factor prediction branch networks. The prediction performance verification module is used to verify the prediction performance of the performance factor prediction branch network set, and determine the branch network weight factor information based on the branch network performance verification results. The weighted fusion module is used to perform weighted fusion and test optimization on the efficacy factor prediction branch network set based on the branch network weight factor information, and to construct a reagent efficacy prediction channel. The dynamic optimization module includes: The efficiency optimization strategy rule base construction module is used to construct an efficiency optimization strategy rule base. The efficiency optimization strategy rule base includes efficiency problem triggering conditions and corresponding efficiency optimization strategies. Specifically, the efficiency optimization strategies are optimization strategies for reagent storage environment and detection parameters. The optimization strategy matching module is used to perform performance problem triggering and optimization strategy matching on the performance optimization strategy rule base based on the target reagent performance information, so as to obtain the target performance optimization strategy; An optimized threshold parsing module is used to perform optimized threshold parsing on the target reagent efficacy information based on the target efficacy optimization strategy, and to construct the efficacy optimization strategy space; The global optimization module is used to perform global optimization on the performance optimization strategy space according to the performance optimization effect evaluation index set, and determine the target performance optimization strategy parameters; The reagent dynamic strategy control module is used to perform strategy optimization feedback on the storage environment and usage detection parameters based on the target performance optimization strategy parameters, obtain reagent performance feedback parameters, and use a PID controller to perform reagent dynamic strategy control based on the reagent performance feedback parameters.

2. The efficacy detection system for the pancreatic elastase 1 detection reagent as described in claim 1, characterized in that, The data mining module includes: The data preprocessing module is used to identify and clean and filter abnormal data in the performance-related data according to data application standards to obtain standard performance-related data. The influencing factor extraction module is used to extract influencing factors from the standard efficacy correlation data to obtain a set of correlation influencing factors and a corresponding reagent efficacy dataset. The correlation coefficient calculation module is used to calculate the correlation coefficient set of each influencing factor in the set of related influencing factors and the set of influencing factors in the reagent efficacy dataset, respectively. The key factor screening module is used to screen the set of related influencing factors based on the set of correlation coefficients of the influencing factors, so as to obtain the set of key factors for reagent efficacy.

3. The efficacy detection system for the pancreatic elastase 1 detection reagent as described in claim 1, characterized in that, The data mining module includes: The correlation feature acquisition module is used to obtain a correlation feature set of efficacy factors based on the set of key factors of reagent efficacy; The feature tree construction module is used to perform cascade analysis and feature tree construction on the performance factor associated feature set respectively, to obtain a set of cascaded feature trees for performance factors; The association data mining module is used to perform association data mining based on the set of cascaded feature trees of the efficacy factors to obtain a dataset of reagent efficacy factors.

4. The efficacy detection system for the pancreatic elastase 1 detection reagent as described in claim 3, characterized in that, The correlation data mining module includes: The data volume requirement hierarchy determination module is used to determine the data volume requirement hierarchy of performance factors based on the reagent performance prediction target. The association depth analysis module is used to perform association depth analysis on the cascaded feature tree set of performance factors according to the hierarchical data volume requirements of the performance factors, and to obtain the feature tree depth of the performance factors; The feature data mining module is used to perform feature data mining based on the depth of the feature tree of the efficacy factors to obtain the dataset of the reagent efficacy factors.

5. The efficacy detection system for the pancreatic elastase 1 detection reagent as described in claim 1, characterized in that, The weighted fusion module includes: The initial performance prediction channel acquisition module is used to perform weighted fusion of the performance factor prediction branch network set based on the branch network weight factor information to obtain the initial performance prediction channel. The test analysis module is used to perform test analysis on the initial performance prediction channel using the performance test dataset to obtain the channel prediction effect parameters; The iterative optimization module is used to select a channel optimizer based on the channel prediction effect parameters, and to iteratively optimize the initial efficacy prediction channel based on the channel optimizer to construct the reagent efficacy prediction channel.

6. A method for detecting the efficacy of a pancreatic elastase 1 assay reagent, characterized in that, The method is implemented using the efficacy detection system of the pancreatic elastase 1 detection reagent according to any one of claims 1-5, and the method includes: Real-time monitoring and acquisition of efficacy correlation data for pancreatic elastase 1 detection reagent, wherein the efficacy correlation data includes environmental parameters, production batch number and instrument parameters; Key factors are extracted from the efficacy-related data to obtain a set of key factors for reagent efficacy, and a dataset of reagent efficacy factors is obtained by mining based on the set of key factors for reagent efficacy. The reagent efficacy factor dataset is trained, fused, and optimized using a time series network structure to build a reagent efficacy prediction channel. Based on the reagent efficacy prediction channel, the efficacy-related data is predicted and analyzed to output target reagent efficacy information. Based on the target reagent efficacy information, an efficacy optimization strategy space is constructed, and based on the efficacy optimization strategy space, the storage environment and detection parameters of the pancreatic elastase 1 detection reagent are dynamically optimized.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the efficacy detection method of the pancreatic elastase 1 detection reagent as described in claim 6.

Images