Data analysis system for prediction of respiratory diseases in children

Abstract

The application discloses a data analysis system for children respiratory disease prediction, and relates to the technical field of health data analysis.The data collection platform comprises a data cleaning model for collecting heterogeneous data of children respiratory disease; the data cleaning model identifies data inconsistency of children respiratory disease, analyzes unstructured clinical patient feedback, and extracts semantic information; a prediction model construction module is used for constructing a hybrid prediction model based on medical environment data and semantic information; a scheme development module is used for inputting the respiratory disease condition of children into the prediction model construction module, analyzing the influence of different climate changes on air quality, and customizing a prevention and treatment scheme.The application realizes real-time early warning of air quality, helps parents and medical institutions to take preventive measures in time, reduces the incidence risk of children respiratory disease, improves the intelligent level of prediction, and provides solid technical support for personalized health management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of health data analysis, in particular to a data analysis system for predicting respiratory diseases in children. BACKGROUND

[0002] At present, with the acceleration of population growth and urbanization, the incidence of children's health problems, especially respiratory diseases, has attracted widespread attention. Modern families pay more and more attention to children's health. In addition to environmental pollution and climate change, the incidence of respiratory diseases in children such as asthma and pneumonia has increased. In order to cope with this challenge, the community actively seeks innovative solutions to improve the efficiency of disease prevention and treatment. Health informatics, as a new discipline, uses advanced information and communication technology to manage and analyze complex medical data, providing strong support for public health and personal health management. Especially in the context of rapid development of big data and artificial intelligence technology, the application potential of health informatics has been unprecedentedly released, becoming an important means to improve the quality of medical services and enhance patient experience.

[0003] Currently, the existing prediction method of respiratory diseases in children mainly relies on the experience of doctors and limited historical case data, which often leads to insufficient accuracy and timeliness of diagnosis. Traditionally, the data collected by medical institutions is mainly structured data. Due to the lack of effective data analysis tools and technology, a large amount of valuable medical data has not been fully mined, thereby limiting the understanding of disease development rules and the improvement of early warning ability. These factors jointly restrict the effect of existing technology in actual application, especially when facing complex and variable respiratory diseases in children, the traditional prediction method is not up to the task. SUMMARY

[0004] In view of the problems existing in the prior art data analysis system for predicting respiratory diseases in children, the present application is proposed.

[0005] In view of the problems existing in the prior art data analysis system for predicting respiratory diseases in children, the present application is proposed.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a data analysis system for predicting respiratory diseases in children, comprising a data collection platform for collecting heterogeneous data of respiratory diseases in children; the data collection platform comprises a data cleaning model, the data cleaning model comprises identifying data inconsistency of respiratory diseases in children, analyzing unstructured clinical patient feedback, and extracting semantic information; collecting medical environment data based on the extracted semantic information;

[0008] a prediction model construction module for constructing a hybrid prediction model based on the medical environment data and the semantic information; the hybrid prediction model is used to capture long-term dependencies of the medical environment data and the semantic information and share knowledge; a scheme development module for inputting the respiratory disease condition of children into the prediction model construction module, analyzing the influence of different climate changes on air quality, and customizing a prevention and treatment scheme.

[0009] As a preferred scheme of the data analysis system for predicting respiratory diseases in children, the data collection platform for constructing respiratory diseases in children comprises a multi-source heterogeneous data integration module, which obtains structured electronic medical record detection results, the multi-source heterogeneous data integration module comprises an adapter for obtaining characteristics of data sources of respiratory diseases in children,

[0010] converting the data collected by the data collection platform into an internal format unified with the characteristics of the data sources, establishing a rule engine in the internal format, detecting inconsistency of the data in the internal format, using a machine learning model to identify and mark abnormal data points, triggering an artificial review process for abnormal data points, and retaining and focusing on normal data points in subsequent analysis.

[0011] As a preferred scheme of the data analysis system for predicting respiratory diseases in children, the integration of heterogeneous data comprises constructing an adaptive classification architecture according to a machine learning model, and classifying the data according to its inconsistency as structured, semi-structured or unstructured;

[0012] the structured data comprises using SQL query language for retrieval, optimizing query through indexing, and when the column title or row data of inconsistent data is queried, the first step is to judge whether it is structured data;

[0013] using a rule engine to check whether each column title or each row of row data conforms to the normal data type, and when all column titles or row data are successfully verified, it is confirmed as structured data;

[0014] the unstructured data comprises using SQL query language for retrieval, and no pre-defined format or structure is found, using a machine learning model to classify the unstructured data, and adjusting the classification results in combination with structured or semi-structured data;

[0015] The semi-structured includes identifying label attributes within the inconsistency of the data by using natural language processing technology, constructing a semantic graph based on the identified label attributes, showing the association between the data, and enabling the machine learning model to learn the semi-structured data pattern through the graph;

[0016] When encountering different data patterns, the known patterns are compared to predict whether they belong to semi-structured data.

[0017] As a preferred scheme of the data analysis system for predicting children's respiratory diseases, wherein: the analysis of unstructured clinical patient feedback includes applying a rule engine to match the classification results of data inconsistency, and checking whether different data patterns have the same label attributes;

[0018] The machine learning model matches the labels in different data patterns, and when both data patterns contain AA labels, the machine learning model considers that they have the same label attributes, at this time, the machine learning model generates a matching report to prompt the user to detect;

[0019] For uncertain matching results, the machine learning model prompts the user to manually match, and the user confirms or modifies the matching result by selecting the operation;

[0020] When the user selects yes, the machine learning model answers the matching question, and when the user selects no, the machine learning model dynamically adjusts the matching rules, at this time, the user points out the misclassification of the label, and the machine learning model will update the rules;

[0021] The extraction of semantic information includes preliminary cleaning of unstructured clinical patient feedback, removing redundant information, using natural language processing technology to segment the unstructured clinical patient feedback, and designing an interactive segmentation interface.

[0022] As a preferred scheme of the data analysis system for predicting children's respiratory diseases, wherein: the collection of medical environment data based on the extracted semantic information includes feature extraction of the interactive segmentation interface, and the feature extraction includes time sequence features, spatial features and text features;

[0023] The time sequence feature includes extracting the time stamp and periodic change feature of the interactive segmentation interface, and capturing the time dependence of the feature extraction of the interactive segmentation interface;

[0024] The spatial feature includes extracting the geographic location and environmental parameter feature of the interactive segmentation interface, and analyzing the differences between different regions;

[0025] The text features include extracting word frequency, sentence length, and punctuation usage features from unstructured text, and using NLP technology to identify medical terms and symptom descriptions;

[0026] The extracted semantic information is associated with medical environment data, and the patient's symptom descriptions are matched with laboratory test results and environmental parameters to form a comprehensive health record. A knowledge graph in the medical field is then constructed based on the comprehensive health record.

[0027] The design of the hybrid prediction model involves using learning algorithms to capture long-term dependencies in time series data, and combining learning algorithms and machine learning models to form a hybrid prediction model.

[0028] The transfer learning technique of the machine learning model includes bringing the hybrid prediction model into a distributed training framework, allocating the model's training tasks to reinforcement learning mechanism computing nodes to accelerate the training process, summarizing the results through a communication mechanism, and optimizing the hybrid prediction model after knowledge sharing.

[0029] As a preferred embodiment of the data analysis system for predicting respiratory diseases in children according to the present invention, the customized prevention and treatment plan includes identifying risk factors through an optimized hybrid prediction model, wherein the risk factors include air quality identified based on collected medical environment data, and the air quality includes climate pattern identification and air quality impact assessment.

[0030] The climate pattern recognition includes integrating meteorological data, air quality data, and geographic information data to build a multi-source data fusion platform and using machine learning models to capture the relationship between climate change and air quality.

[0031] The method of capturing the relationship between climate change and air quality includes using time series analysis and optimized hybrid prediction models to predict future climate patterns and changes in air quality.

[0032] As a preferred embodiment of the data analysis system for predicting respiratory diseases in children according to the present invention, the analysis of the impact of different climate changes on air quality includes introducing a weather event prediction model to assess the impact of events on air quality;

[0033] When a weather event prediction model is introduced, an early warning for air quality is issued by combining the weather event prediction model with air quality data and geographic information data.

[0034] When the PM2.5 concentration in air quality data rises, an early warning mechanism will be triggered. Weather event prediction models will monitor future climate patterns and air quality changes, and complete data analysis for predicting respiratory diseases in children.

[0035] Secondly, this invention provides a data analysis method for predicting respiratory diseases in children, comprising: constructing a data collection platform for children's respiratory diseases, identifying data inconsistencies in children's respiratory diseases, parsing unstructured clinical patient feedback, and extracting semantic information; collecting medical environment data based on the extracted semantic information, designing a hybrid prediction model, and optimizing the hybrid prediction model after knowledge sharing by introducing a reinforcement learning mechanism; and in the optimized knowledge-sharing hybrid prediction model, customizing prevention and treatment plans according to the child's respiratory disease condition.

[0036] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the steps of the data analysis system for predicting respiratory diseases in children described above.

[0037] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any of the steps of the data analysis system for predicting respiratory diseases in children described above.

[0038] The beneficial effects of this invention are as follows: It significantly improves prediction accuracy and personalized prevention effectiveness. By constructing a multi-source heterogeneous data integration platform, it integrates structured, semi-structured, and unstructured data, solving the data inconsistency problem and ensuring data integrity and reliability. Natural language processing technology is used to analyze clinical feedback and extract semantic information, enriching the data depth. The designed hybrid prediction model combines deep learning to capture long-term dependencies and achieves knowledge sharing through transfer learning, improving the model's generalization ability. A reinforcement learning mechanism is introduced to optimize the model, enabling it to adaptively adjust and ensuring the real-time nature and accuracy of prediction results. Furthermore, by analyzing the impact of climate change on air quality and combining it with a weather event prediction model, it achieves real-time air quality warnings, helping parents and medical institutions take timely preventative measures and reduce the risk of respiratory diseases in children. Overall, this invention improves the intelligence level of prediction and provides solid technical support for personalized health management. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0040] Figure 1 This is a flowchart of a data analysis system for predicting respiratory diseases in children.

[0041] Figure 2 This is an internal structure diagram of a computer device used for a data analysis system to predict respiratory diseases in children. Detailed Implementation

[0042] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0043] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0045] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0046] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0047] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0048] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a data analysis system for predicting respiratory diseases in children, including:

[0049] Data acquisition platform, prediction model building module, and solution formulation module;

[0050] A data acquisition platform is used to collect heterogeneous data on pediatric respiratory diseases; it includes a data cleaning model, which includes identifying data inconsistencies in pediatric respiratory diseases, parsing unstructured clinical patient feedback, and extracting semantic information; and collecting medical environment data based on the extracted semantic information.

[0051] The prediction model building module is used to construct a hybrid prediction model based on the medical environment data and the semantic information; and to use the hybrid prediction model to capture the long-term dependencies between the medical environment data and the semantic information and to share knowledge.

[0052] The plan formulation module is used to input the child's respiratory disease information into the prediction model construction module, analyze the impact of different climate changes on air quality, and customize prevention and treatment plans.

[0053] Specifically, the data acquisition platform includes designing a multi-source heterogeneous data integration module to obtain structured electronic medical record test results. The design of the multi-source heterogeneous data integration module includes using an adapter to obtain the data source characteristics of children's respiratory diseases.

[0054] The data collected by the data acquisition platform is converted into an internal format that is consistent with the characteristics of the data source. A rule engine is built in the internal format to detect inconsistencies in the data. Machine learning models are used to identify and mark abnormal data points. For abnormal data points, a manual review process is triggered, while normal data points are retained and monitored in subsequent analysis.

[0055] The integration of heterogeneous data includes building an adaptive classification architecture based on machine learning models and classifying the data as structured, semi-structured, or unstructured based on inconsistencies.

[0056] Structured data includes using SQL query language for retrieval, optimizing queries through indexes, and when inconsistent column headers or row data are found in the query, the first step is to determine whether the data is structured.

[0057] The rules engine is used to check whether the column headers of each column or the row data of each row conform to the normal data type. When all column headers or row data are successfully validated, the data is confirmed to be structured data.

[0058] Unstructured data includes data retrieved using SQL query language, where no predefined format or structure was found. Unstructured data is then classified using machine learning models, and the classification results are adjusted by combining structured or semi-structured data.

[0059] Semi-structured data processing involves using natural language processing techniques to identify label attributes within data inconsistencies, constructing semantic graphs based on the identified label attributes to display the relationships between data, and using the graphs to enable machine learning models to learn semi-structured data patterns.

[0060] When encountering different data patterns, the system can predict whether the data belongs to semi-structured data by comparing it with known patterns.

[0061] Furthermore, parsing unstructured clinical patient feedback involves applying a rule engine to match data based on inconsistencies in the classification results and checking whether different data patterns have the same label attributes.

[0062] The machine learning model matches labels in different data patterns. When two data patterns contain AA labels, the machine learning model considers them to have the same label attributes. At this time, the machine learning model generates a matching report to prompt the user to perform the detection.

[0063] For uncertain matching results, the machine learning model will prompt the user to perform a manual matching. The user can then confirm or modify the matching result by making a selection.

[0064] When the user selects "yes", the machine learning model answers the matching question. When the user selects "no", the machine learning model dynamically adjusts the matching rules. At this time, if the user points out the misclassification of the label, the machine learning model will update the rules.

[0065] Extracting semantic information includes preliminary cleaning of unstructured clinical patient feedback, removing redundant information, using natural language processing technology to segment unstructured clinical patient feedback into words, and designing an interactive word segmentation interface.

[0066] The prediction model building module collects medical environment data based on extracted semantic information, designs a hybrid prediction model, combines deep learning algorithms to capture long-term dependencies in time series data, adopts transfer learning technology to enable knowledge sharing in the hybrid prediction model, and optimizes the hybrid prediction model after knowledge sharing by introducing a reinforcement learning mechanism.

[0067] Among them, collecting medical environment data based on extracted semantic information includes feature extraction of the interactive word segmentation interface, and feature extraction includes time series features, spatial features and text features;

[0068] Time series features include extracting timestamps and periodic change features from the interactive word segmentation interface, and capturing the time dependency of feature extraction from the interactive word segmentation interface;

[0069] Spatial features include extracting geographical location and environmental parameter features from the interactive word segmentation interface, and analyzing the differences between different regions;

[0070] Text features include extracting word frequency, sentence length, and punctuation usage features from unstructured text, and using NLP technology to identify medical terms and symptom descriptions;

[0071] The extracted semantic information is associated with medical environment data, and the patient's symptom descriptions are matched with laboratory test results and environmental parameters to form a comprehensive health record. A knowledge graph in the medical field is then constructed based on the comprehensive health record.

[0072] The design of the hybrid prediction model involves using learning algorithms to capture long-term dependencies in time series data, and combining learning algorithms and machine learning models to form a hybrid prediction model.

[0073] Transfer learning techniques for machine learning models include bringing hybrid prediction models into a distributed training framework, distributing the model's training tasks to reinforcement learning mechanism computing nodes to accelerate the training process, and summarizing the results through communication mechanisms to optimize the hybrid prediction model after knowledge sharing.

[0074] The solution development module, within the optimized knowledge-sharing hybrid prediction model, customizes prevention and treatment plans based on children's respiratory disease conditions, analyzes the impact of different climate changes on air quality, and completes data analysis for predicting children's respiratory diseases.

[0075] Among them, the customized prevention and treatment plan includes identifying risk factors through an optimized hybrid prediction model. Risk factors include air quality identification based on collected medical environmental data. Air quality includes climate pattern identification and air quality impact assessment.

[0076] Climate pattern recognition involves integrating meteorological data, air quality data, and geographic information data to build a multi-source data fusion platform and using machine learning models to capture the relationship between climate change and air quality.

[0077] Capturing the relationship between climate change and air quality involves using time series analysis and optimized hybrid forecasting models to predict future climate patterns and changes in air quality.

[0078] Furthermore, the analysis of the impacts of different climate changes on air quality includes the introduction of weather event prediction models to assess the impact of events on air quality;

[0079] When a weather event prediction model is introduced, an early warning for air quality is issued by combining the weather event prediction model with air quality data and geographic information data.

[0080] When the PM2.5 concentration in air quality data rises, an early warning mechanism will be triggered. Weather event prediction models will monitor future climate patterns and air quality changes, and complete data analysis for predicting respiratory diseases in children.

[0081] Furthermore, in the optimized knowledge-sharing hybrid prediction model, the system customizes prevention and treatment plans based on children's respiratory illnesses and analyzes the impact of climate change on air quality. Specifically, when the average PM2.5 concentration over the past week is detected to be 40 micrograms per cubic meter, exceeding the World Health Organization's recommended safety standard of 25 micrograms per cubic meter, the system advises parents to reduce their children's outdoor activities and use air purifiers. Simultaneously, the system detects an increase in sulfur dioxide and nitrogen oxide concentrations in the area, further exacerbating the risk of air pollution. For children with a history of asthma, when the PM2.5 concentration exceeds 35 micrograms per cubic meter, the system reminds parents to prepare necessary medications and take their children for regular checkups. The system integrates data from meteorological stations, air quality monitoring stations, and geographic information systems to construct a multi-source data fusion platform. It monitors in real time temperature, humidity, wind speed, precipitation, and indicators such as PM2.5, PM10, ozone, sulfur dioxide, and nitrogen oxides. Through time series analysis, the system predicts that the PM2.5 concentration in a certain city will gradually increase over the next seven days, especially on the fourth and fifth days, when the PM2.5 concentration may reach over 50 micrograms per cubic meter, exceeding safety standards. Furthermore, the system predicts that temperatures will drop during the same period, potentially leading to increased demand for coal-fired heating, further impacting air quality. Data on respiratory illnesses in children and the impact on air quality are shown in Table 1 below.

[0082] Table 1. Data on the impact of respiratory diseases in children on air quality

[0083] Data type Value Unit Description PM2.5 average concentration 40 Micrograms per cubic meter The average concentration of PM2.5 in a city over the past week, exceeding the safe standard recommended by the World Health Organization World Health Organization safe standard 25 Micrograms per cubic meter The safe standard for PM2.5 recommended by the World Health Organization PM2.5 concentration threshold 35 Micrograms per cubic meter When the PM2.5 concentration exceeds this value, parents are advised to take preventive measures Predicted PM2.5 concentration 50 Micrograms per cubic meter The system predicts the level that the PM2.5 concentration in a city may reach on the fourth and fifth days within the next seven days Prediction period 7 Days The system predicts the change in air quality within the next seven days Temperature change Decrease - The system predicts that the temperature will decrease during the same period, which may increase the demand for coal heating

[0084] Table 1 shows the specific data involved in the system's analysis of children's respiratory diseases and the impact of air quality. This data provides a scientific basis for developing prevention and treatment plans and helps parents and medical institutions take timely and effective protective measures.

[0085] In a preferred embodiment, a data analysis system for predicting pediatric respiratory diseases includes a construction module that builds a data collection platform for pediatric respiratory diseases, identifies data inconsistencies, analyzes unstructured clinical patient feedback, and extracts semantic information; an optimization module that collects medical environment data based on the extracted semantic information, designs a hybrid prediction model, and optimizes the knowledge-sharing hybrid prediction model by introducing a reinforcement learning mechanism; and a prediction module that, based on the optimized knowledge-sharing hybrid prediction model, customizes prevention and treatment plans according to the child's respiratory disease condition.

[0086] The computer device can be a terminal, and it includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. Its internal structure diagram can be as follows: Figure 2 As shown, the processor of this computer device provides computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The communication interface of this computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be an LCD screen or an e-ink display screen. The input device of this computer device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad located on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0087] In summary, this invention significantly improves prediction accuracy and personalized prevention effectiveness. By constructing a multi-source heterogeneous data integration platform, it integrates structured, semi-structured, and unstructured data, solving the data inconsistency problem and ensuring data integrity and reliability. Natural language processing technology is used to analyze clinical feedback and extract semantic information, enriching the data depth. The designed hybrid prediction model combines deep learning to capture long-term dependencies and achieves knowledge sharing through transfer learning, improving the model's generalization ability. A reinforcement learning mechanism is introduced to optimize the model, enabling it to adaptively adjust and ensuring the real-time nature and accuracy of prediction results. Furthermore, by analyzing the impact of climate change on air quality and combining it with a weather event prediction model, real-time air quality warnings are achieved, helping parents and medical institutions take timely preventative measures and reduce the risk of respiratory diseases in children. Overall, this invention enhances the intelligence level of prediction and provides solid technical support for personalized health management.

[0088] Example 2, refer to Figure 1This is the second embodiment of the present invention, which provides a data analysis system for predicting respiratory diseases in children. In order to verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0089] The extracted semantic information was correlated with medical environment data to form a comprehensive health record, which included symptom descriptions of 500 patients, 300 laboratory test results, and 200 environmental parameters. This data was used to construct a knowledge graph in the medical field. The designed hybrid prediction model used the LSTM algorithm to capture long-term dependencies in time-series data and combined it with the XGBoost machine learning model to form a hybrid prediction model. During training, 10,000 historical data points and 2,000 validation data points were used, and the model's accuracy reached 92%. Transfer learning technology was used to bring the hybrid prediction model into a distributed training framework, distributing the model's training task across 10 computing nodes, with each node processing 1,000 data points, thus accelerating the training process. The results from each node were aggregated through a communication mechanism, and the accuracy of the optimized hybrid prediction model on the test set was improved to 95%. Experimental results show that this method can effectively improve the accuracy and efficiency of predicting respiratory diseases in children. The experimental data of this invention are shown in Table 2 below:

[0090] Table 2 Experimental Data of the Invention

[0091] Stage / component Description Key data Data association and comprehensive health record Associate extracted semantic information with medical environment data to form a comprehensive health record Symptom descriptions of 500 patients, 300 laboratory test results, and 200 environmental parameters Knowledge graph construction Construct a knowledge graph in the medical field based on the comprehensive health record - Hybrid prediction model design Use LSTM algorithm to capture long-term dependencies in time series data, and combine XGBoost machine learning model to form a hybrid prediction model LSTM + XGBoost Training data Use historical data for model training 10,000 historical data Validation data Use validation data to evaluate model performance 2,000 validation data Initial accuracy The accuracy of the model before optimization 92% Distributed training framework Migrate the hybrid prediction model to a distributed training framework using transfer learning technology 10 computing nodes, each processing 1,000 data Communication mechanism Aggregate the results of each node through the communication mechanism - Optimized accuracy The accuracy of the optimized hybrid prediction model on the test set 95%

[0092] Table 2 summarizes the key steps and related data from data association to final model optimization, making the information more intuitive and readable. Since the original text did not mention the specific amount of test data, this item is not included in the table. If more detailed information is needed, the table content can be further refined or additional columns can be added to cover more details. A comparison of this invention with existing technologies is shown in Table 3 below:

[0093] Table 3 Comparison of the present invention with the prior art

[0094] Comparison item Prior art The invention Advantages of the invention Data collection and integration Dependence on a single data source makes it difficult to integrate multi-source heterogeneous data; data cleaning and preprocessing process is complex and inefficient A multi-source heterogeneous data integration module is constructed, which can efficiently convert and unify data formats, automatically detect and mark abnormal data points Supports efficient integration of multi-source heterogeneous data, improving data quality and processing efficiency Semantic information extraction It is difficult to effectively parse unstructured text such as clinical patient feedback, resulting in insufficient utilization of semantic information Apply rule engine and machine learning model to match label attributes, and use natural language processing technology for word segmentation and feature extraction Improves the understanding of unstructured text, enhancing the accuracy and usability of semantic information Prediction model design Mainly rely on traditional statistical methods or single algorithm, difficult to capture long-term dependencies in time series Design a hybrid prediction model that combines LSTM and XGBoost, which can more effectively capture long-term dependencies in time series data The design of the hybrid model improves the prediction accuracy, especially suitable for respiratory disease prediction with time dependence Transfer learning application Lack of effective knowledge sharing mechanism, low model training efficiency The introduction of transfer learning technology and the mixed prediction model into a distributed training framework has accelerated the training process Through transfer learning, knowledge sharing is achieved, significantly shortening the model training time and improving the training efficiency Computing resource utilization Single-node training is usually used, resulting in low computing resource utilization The training task is distributed to 10 computing nodes, with each node processing 1000 data, and the results are aggregated through a communication mechanism The distributed training framework significantly improves the utilization of computing resources and speeds up the model optimization Prediction accuracy The accuracy rate is usually around 85%, leaving room for improvement The optimized mixed prediction model achieves an accuracy rate of 95% on the test set Significantly improves prediction accuracy, providing a more reliable basis for the prevention and treatment of children's respiratory diseases

[0095] Table 3 illustrates the main differences between this invention and existing technologies, highlighting the significant advantages of this invention in several key areas, including data processing, semantic information extraction, prediction model design, transfer learning application, computing resource utilization, prediction accuracy, and climate change and air quality analysis. These advantages together constitute the core competitiveness of this invention, giving it a clear technological lead in the field of pediatric respiratory disease prediction.

[0096] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A data analysis system for predicting respiratory diseases in children, characterized in that: include: A data acquisition platform for collecting heterogeneous data on children's respiratory diseases; It includes a data cleaning model, which includes identifying data inconsistencies in pediatric respiratory diseases, parsing unstructured clinical patient feedback, and extracting semantic information; and collecting medical environment data based on the extracted semantic information. The prediction model building module is used to construct a hybrid prediction model based on the medical environment data and the semantic information; and to use the hybrid prediction model to capture the long-term dependencies between the medical environment data and the semantic information, and to share knowledge. The plan formulation module is used to input the child's respiratory disease information into the prediction model construction module, analyze the impact of different climate changes on air quality, and customize prevention and treatment plans. The data acquisition platform includes a multi-source heterogeneous data integration module to obtain structured electronic medical record detection results. The multi-source heterogeneous data integration module includes using an adapter to obtain the data source characteristics of children's respiratory diseases. The data collected by the data acquisition platform is converted into an internal format that is consistent with the characteristics of the data source. A rule engine is built in the internal format to detect inconsistencies in the data. Machine learning models are used to identify and mark abnormal data points. For abnormal data points, a manual review process is triggered. For normal data points, they are retained and monitored in subsequent analysis. The integration of heterogeneous data includes building an adaptive classification architecture based on a machine learning model, and classifying the data into structured, semi-structured, or unstructured categories based on data inconsistencies. The structuring process includes using SQL query language for retrieval, optimizing queries through indexes, and when inconsistent column headers or row data are found in the query, the first step is to determine that the data is structured. The rules engine is used to check whether the column headers of each column or the row data of each row conform to the normal data type. When all column headers or row data are successfully verified, the data is confirmed to be structured data. The unstructured data includes retrieval using SQL query language, where no predefined format or structure was found, classification of unstructured data using machine learning models, and adjustment of the classification results by combining structured or semi-structured data. The semi-structured approach includes using natural language processing technology to identify label attributes within data inconsistencies, constructing a semantic graph based on the identified label attributes to display the relationships between data, and enabling machine learning models to learn semi-structured data patterns through the graph. When encountering different data patterns, predict whether they belong to semi-structured data by comparing them with known patterns; The parsing of unstructured clinical patient feedback includes applying a rule engine to match data based on inconsistencies in the classification results and checking whether different data patterns have the same label attributes. The machine learning model matches labels in different data patterns. When two data patterns contain AA labels, the machine learning model considers them to have the same label attributes. At this time, the machine learning model generates a matching report to prompt the user to perform the detection. For uncertain matching results, the machine learning model will prompt the user to perform a manual matching. The user can then confirm or modify the matching result by making a selection. When the user selects "yes", the machine learning model answers the matching question. When the user selects "no", the machine learning model dynamically adjusts the matching rules. At this time, if the user points out the misclassification of the label, the machine learning model will update the rules. The extraction of semantic information includes preliminary cleaning of unstructured clinical patient feedback, removal of redundant information, word segmentation of unstructured clinical patient feedback using natural language processing technology, and design of an interactive word segmentation interface. The customized prevention and treatment plan includes identifying risk factors through an optimized hybrid prediction model, including air quality identification based on collected medical environmental data, and air quality including climate pattern identification and air quality impact assessment; The climate pattern recognition includes integrating meteorological data, air quality data, and geographic information data to build a multi-source data fusion platform and using machine learning models to capture the relationship between climate change and air quality. The method of capturing the relationship between climate change and air quality includes using time series analysis and optimized hybrid prediction models to predict future climate patterns and changes in air quality.

2. The data analysis system for predicting respiratory diseases in children as described in claim 1, characterized in that: The collection of medical environment data based on extracted semantic information includes feature extraction of the interactive word segmentation interface, which includes time series features, spatial features, and text features; The time series features include extracting timestamps and periodic change features from the interactive word segmentation interface, and capturing the time dependency of feature extraction from the interactive word segmentation interface. The spatial features include extracting the geographical location and environmental parameter features of the interactive word segmentation interface, and analyzing the differences between different regions; The text features include extracting word frequency, sentence length, and punctuation usage features from unstructured text, and using NLP technology to identify medical terms and symptom descriptions; The extracted semantic information is associated with medical environment data, and the patient's symptom descriptions are matched with laboratory test results and environmental parameters to form a comprehensive health record. A knowledge graph in the medical field is then constructed based on the comprehensive health record. The design of the hybrid prediction model involves using learning algorithms to capture long-term dependencies in time series data, and combining learning algorithms and machine learning models to form a hybrid prediction model. The transfer learning technique of the machine learning model includes bringing the hybrid prediction model into a distributed training framework, allocating the model's training tasks to reinforcement learning mechanism computing nodes to accelerate the training process, summarizing the results through a communication mechanism, and optimizing the hybrid prediction model after knowledge sharing.

3. The data analysis system for predicting respiratory diseases in children as described in claim 2, characterized in that: The analysis of the impacts of different climate changes on air quality includes the introduction of weather event prediction models to assess the impact of events on air quality. When a weather event prediction model is introduced, an early warning for air quality is issued by combining the weather event prediction model with air quality data and geographic information data. When the PM2.5 concentration in air quality data rises, an early warning mechanism will be triggered. Weather event prediction models will monitor future climate patterns and air quality changes, and complete data analysis for predicting respiratory diseases in children.

4. A data analysis method for predicting respiratory diseases in children, based on the data analysis system for predicting respiratory diseases in children as described in any one of claims 1 to 3, characterized in that: include: Build a data collection platform for pediatric respiratory diseases, identify data inconsistencies in pediatric respiratory diseases, analyze unstructured clinical patient feedback, and extract semantic information; Based on the extracted semantic information, medical environment data is collected, a hybrid prediction model is designed, and the hybrid prediction model after knowledge sharing is optimized by introducing a reinforcement learning mechanism. In the optimized knowledge-sharing hybrid prediction model, prevention and treatment plans are customized based on the child's respiratory disease status.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the data analysis system for predicting respiratory diseases in children as described in any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the data analysis system for predicting respiratory diseases in children as described in any one of claims 1 to 3.

Images