Clinical decision support system and method for respiratory tract infectious diseases of children

Abstract

The invention discloses a children respiratory tract infectious disease clinical decision support system and method, and relates to the technical field of health data diagnosis decision, and the method comprises the steps: mining a distributed calculation framework of children respiratory tract infectious diseases; performing multi-source data fusion on the reference data and the distributed computing framework; dynamic monitoring and early warning of a clinical intelligent diagnosis model are established through the internet technology, the health state change of a child patient is tracked, and clinical decision-making of the respiratory tract infectious diseases of the child is completed. According to the method, the comprehensiveness and reliability of data are improved, the health state change of the child patient is tracked in time, early recognition and precise treatment are realized, the data safety and privacy protection are guaranteed, powerful support is provided for clinical decision making, and the diagnosis and treatment level of the child respiratory tract infectious diseases is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health data diagnosis and decision-making, and in particular to a clinical decision support system and method for children's respiratory tract infectious diseases. Background Art

[0002] With the acceleration of the global informatization process, healthcare informatics, as an information and communication technology specifically used to handle or process medical or health data, has become a key force in promoting the modernization and transformation of the medical and health industry. Especially in the field of children's health, facing the challenges brought by common diseases such as respiratory infections, the application of healthcare informatics is particularly important. It not only helps to improve the early identification rate and diagnostic accuracy of diseases, but also provides doctors with more scientific decision-making support through intelligent data analysis, thereby improving the overall medical service level and ensuring the healthy growth of children.

[0003] At present, the existing medical information system still has some shortcomings when dealing with children's respiratory infections. For example, the data island phenomenon is serious, information sharing between different medical institutions is not smooth, diagnosis relies on the personal experience of doctors, lacks systematic data support, and treatment plans are mostly general and fail to fully consider individual differences. Summary of the invention

[0004] In view of the problems existing in the existing clinical decision support system and method for children's respiratory tract infectious diseases, the present invention is proposed.

[0005] Therefore, the purpose of the present invention is to provide a clinical decision support system and method for children's respiratory infectious diseases. In view of the problems of low data quality, low efficiency of multi-source data fusion, and untimely dynamic monitoring and early warning in the prior art clinical data processing and decision support systems for children's respiratory infectious diseases, the present invention adopts a distributed computing framework and an intelligent diagnosis model to improve data quality and the real-time and accuracy of decision-making through data point cleaning, multi-source data fusion and real-time prediction.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a clinical decision support method for children's respiratory tract infectious diseases, which includes: using a clinical decision support system for children's respiratory tract infectious diseases to mine a distributed computing framework for children's respiratory tract infectious diseases; Collect clinical reference data of pediatric respiratory infectious diseases in the external environment, and fuse the reference data with the distributed computing framework for multi-source data; A clinical intelligent diagnosis model is constructed based on the fused multi-source data, and a dynamic monitoring and early warning system for the clinical intelligent diagnosis model is established through Internet technology to track changes in the health status of child patients and complete clinical decision-making for children's respiratory infections.

[0007] As a preferred embodiment of the clinical decision support method for children's respiratory tract infectious diseases of the present invention, wherein: the use of the clinical decision support system for children's respiratory tract infectious diseases includes using a data point cleaning algorithm to remove erroneous data points of clinical symptoms, and unifying the data format of the correct data points of clinical symptoms through standardization processing, and the use of the data point cleaning algorithm includes calculating the number of missing values ​​of clinical characteristics of children's respiratory tract infectious diseases, calculating the proportion of the number of missing values ​​of clinical characteristics of children's respiratory tract infectious diseases according to the number of missing values ​​of characteristics, and calculating the average proportion of the number of missing values ​​of characteristics based on the proportion of the number of missing values ​​of characteristics; The formula for calculating the number of missing values ​​of clinical characteristics of children's respiratory infections is: ; in, Indicates Record No. eigenvalues, represents the indicator function, Indicates a record bar. Indicates the number of missing values ​​of features; When Record No. When the feature value returns 1, there is a missing value; Otherwise, when Record No. When the feature value returns 0, there is no missing value; The formula for the proportion of missing values ​​of the feature is: ; in, represents the proportion of missing values ​​of features, represents the number of missing values ​​of features, Indicates a record bar.

[0008] As a preferred embodiment of the clinical decision support method for children's respiratory tract infectious diseases of the present invention, the formula for the average proportion of the number of missing values ​​of the features is: ; in, represents the average proportion of missing values ​​of features, Indicates the number of features of the record bar, Indicates the proportion of missing values ​​of features; The distributed computing framework for mining children's respiratory tract infection diseases includes calculating the distribution standard deviation of the number of feature missing values ​​according to the average proportion of the number of feature missing values, and the formula for the distribution standard deviation is: ; in, represents the standard deviation of the distribution; The threshold of the distributed computing framework is calculated according to the distribution standard deviation, and the formula of the threshold is: ; in, represents the adjustment factor, represents the threshold of the distributed computing framework; When the number of missing values ​​of features is proportional Greater than threshold When , the number of missing feature values ​​is removed. At this time, the number of missing feature values ​​is the wrong data point of clinical symptoms; When the number of missing values ​​of features is proportional Less than or equal to threshold When , it is the correct data point of the clinical symptoms. At this time, the data format is unified by filling the mean, median and interpolation method of the proportion of the number of missing values ​​of the features.

[0009] As a preferred embodiment of the clinical decision support method for children's respiratory tract infectious diseases of the present invention, wherein: the collection of clinical reference data of children's respiratory tract infectious diseases in the external environment includes processing the data source in the external environment with the unified correct data points, and the processing includes setting user permissions and isolating the data source in the external environment from the unified correct data points; The setting of user permissions includes allowing staff to access unified pediatric patient data at correct data points; Otherwise, when user permissions are not set, only ordinary users are allowed to view the summary data; The isolating the data source in the external environment from the unified correct data point includes preventing the data source in the external environment from affecting the unified correct data point; Based on the unified correct data points, verify the data source in the external environment to check whether the data source in the external environment is consistent with the unified correct data points; When the correct data point after unification increases or decreases, it indicates that there is clinical inconsistency in children's respiratory infectious diseases and is marked as False; Otherwise, when the correct data points after unification are consistent, it means that there is clinical consistency in children's respiratory infectious diseases, which is marked as True. The specific steps of the verification are: When Records When it is equal to True, the verification passes; Otherwise, when Records If it is equal to False, the verification fails.

[0010] As a preferred solution of the clinical decision support method for children's respiratory tract infectious diseases of the present invention, wherein: the multi-source data fusion of the reference data and the distributed computing framework includes using the task scheduler of the distributed computing framework to allocate the data processing tasks of the reference data to different distributed computing framework nodes; Design the synchronization mechanism of reference data according to the distributed computing framework nodes, keep the distributed computing framework nodes consistent with the correct data points after unification, and set up the exception handling synchronization mechanism; When the distributed computing framework nodes are consistent with the correct data points after unification, the multi-source data fusion is normal; When the distributed computing framework node is inconsistent with the correct data point after unification, the multi-source data fusion is abnormal. At this time, the administrator is notified to conduct an investigation.

[0011] As a preferred solution of the clinical decision support method for pediatric respiratory tract infectious diseases of the present invention, wherein: the construction of a clinical intelligent diagnosis model based on the fused multi-source data includes preprocessing the fused multi-source data, and the preprocessing includes feature selection and feature scaling of the multi-source data fusion; The feature selection includes selecting the diagnostic results of clinical decision-making for children's respiratory tract infectious diseases, and the specific formula is: ; in, represents the selected feature set, The diagnostic result eigenvalues, represents the feature selection threshold; The feature scaling includes scaling the first The eigenvalues ​​are scaled to a uniform range and unified using a normalized calculation method. The normalized calculation formula is: ; in, represents the scaled normalized eigenvalue, The diagnostic result The mean of the eigenvalues, The diagnostic result The standard deviation of the eigenvalues.

[0012] As a preferred solution of the clinical decision support method for children's respiratory tract infectious diseases of the present invention, wherein: the dynamic monitoring and early warning of establishing a clinical intelligent diagnosis model includes using the clinical intelligent diagnosis model to predict the scaled normalized eigenvalues, tracking the changes in the health status of the child patient, and judging whether the child patient has an infection risk; The prediction result calculation formula for predicting the scaled normalized eigenvalue is: ; in, Indicates at time The prediction results, Indicates time, Indicates the real time at this moment; when The prediction result of real time is greater than the feature selection threshold When the pediatric respiratory infection disease clinical decision support system issues an early warning, the clinical decision support system for pediatric respiratory infection disease is completed according to the set user authority; Otherwise, when The prediction result of real time is less than or equal to the feature selection threshold When the child is monitored, the changes in health status of the child patient can be tracked.

[0013] In a second aspect, the present invention provides a clinical decision support system for children's respiratory tract infectious diseases, which includes: an application module, which uses the clinical decision support system for children's respiratory tract infectious diseases to mine a distributed computing framework for children's respiratory tract infectious diseases; The acquisition module collects clinical reference data of children's respiratory infectious diseases in the external environment and integrates the reference data with the distributed computing framework for multi-source data fusion; The decision-making module builds a clinical intelligent diagnosis model based on the integrated multi-source data, establishes dynamic monitoring and early warning of the clinical intelligent diagnosis model through Internet technology, tracks changes in the health status of child patients, and completes clinical decision-making for children's respiratory infections.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, any step of the above-mentioned clinical decision support method for children's respiratory infectious diseases is implemented.

[0015] In a fourth aspect, 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, any step of the above-mentioned clinical decision support method for children's respiratory infectious diseases is implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention effectively improves the quality and processing efficiency of clinical data on children's respiratory infectious diseases by using a distributed computing framework and an intelligent diagnostic model, removes erroneous data points through a data point cleaning algorithm to ensure the accuracy and consistency of the data, integrates information from different data sources through multi-source data fusion technology to improve the comprehensiveness and reliability of the data, and timely tracks changes in the health status of child patients through real-time dynamic monitoring and early warning to achieve early identification and precise treatment. In addition, the present invention also designs a strict user authority management and data verification mechanism to ensure data security and privacy protection, provides strong support for clinical decision-making, and significantly improves the diagnosis and treatment level of children's respiratory infectious diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them: Figure 1 A specific flow chart of a clinical decision support method for children's respiratory tract infectious diseases provided by one embodiment of the present invention.

[0018] Figure 2 A decision diagram of a clinical decision support method for children's respiratory tract infectious diseases provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] 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 term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0022] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0023] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0024] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] Example 1: Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a clinical decision support method for children's respiratory tract infectious diseases, including: S1: Use the clinical decision support system for children's respiratory infections to explore the distributed computing framework for children's respiratory infections.

[0026] Among them, the use of the clinical decision support system for children's respiratory tract infectious diseases includes using a data point cleaning algorithm to remove erroneous data points of clinical symptoms, unifying the data format of correct data points of clinical symptoms through standardization processing, and using the data point cleaning algorithm includes calculating the number of missing values ​​of clinical characteristics of children's respiratory tract infectious diseases, calculating the proportion of the number of missing values ​​of clinical characteristics of children's respiratory tract infectious diseases based on the number of missing values ​​of characteristics, and calculating the average proportion of the number of missing values ​​of characteristics based on the proportion of the number of missing values ​​of characteristics; The formula for calculating the number of missing values ​​for clinical features of childhood respiratory infections is: ; in, Indicates Record No. eigenvalues, represents the indicator function, Indicates a record bar. Indicates the number of missing values ​​of features; When Record No. When the feature value returns 1, there is a missing value; Otherwise, when Record No. When the feature value returns 0, there is no missing value; The formula for the proportion of missing values ​​of the feature is: ; in, represents the proportion of missing values ​​of features, represents the number of missing values ​​of features, Indicates a record bar.

[0027] S1.1: The formula for the average proportion of missing values ​​is: ; in, represents the average proportion of missing values ​​of features, Indicates the number of features of the record bar, Indicates the proportion of missing values ​​of features; The distributed computing framework for mining children's respiratory infections includes calculating the distribution standard deviation of the number of missing values ​​of features based on the average proportion of the number of missing values ​​of features. The formula for the distribution standard deviation is: ; in, represents the standard deviation of the distribution; The threshold of the distributed computing framework is calculated based on the distribution standard deviation. The formula for the threshold is: ; in, represents the adjustment factor, represents the threshold of the distributed computing framework; When the number of missing values ​​of features is proportional Greater than threshold When , the number of missing feature values ​​is removed. At this time, the number of missing feature values ​​is the wrong data point of clinical symptoms; When the number of missing values ​​of features is proportional Less than or equal to threshold When , it is the correct data point of the clinical symptoms. At this time, the data format is unified by filling the mean, median and interpolation method of the proportion of the number of missing values ​​of the features.

[0028] Furthermore, when calculating the threshold of the distributed computing framework, the standard deviation of the distribution of the number of missing values ​​of the features is first calculated based on the average proportion of the number of missing values ​​of the features, and then the threshold is determined based on the distribution standard deviation and an adjustment coefficient. Specifically, the threshold is calculated by adding the average proportion of the number of missing values ​​of the features to the adjustment coefficient and multiplying the distribution standard deviation. If the missing value proportion of a feature is greater than the threshold, the missing value of the feature is considered to be an erroneous data point and needs to be removed. Conversely, if the missing value proportion of the feature is less than or equal to the threshold, the missing value of the feature is considered to be a correct data point, and the data format can be unified by filling the mean, median or interpolation filling method. The adjustment coefficient is usually set to a value between 0 and 3 to adapt to different data distribution situations. For example, if the average proportion of the number of missing values ​​of the feature is 0.1, the distribution standard deviation is 0.05, and the adjustment coefficient is 2, the calculated threshold is 0.2, which means that when the missing value proportion of a feature is greater than 0.2, the missing value of the feature will be regarded as an erroneous data point and removed; otherwise, the data format will be unified by the filling method.

[0029] S2: Collect clinical reference data on children's respiratory infectious diseases in the external environment, and fuse the reference data with the distributed computing framework for multi-source data integration.

[0030] Among them, collecting clinical reference data of children's respiratory infectious diseases in the external environment includes processing the data source in the external environment with the unified correct data points, and the processing includes setting user permissions and isolating the data source in the external environment from the unified correct data points; Setting user permissions includes allowing staff to access pediatric patient data at the correct data point after harmonization; Otherwise, when user permissions are not set, only ordinary users are allowed to view the summary data; Isolating the data source in the external environment from the correct data point after unification includes preventing the data source in the external environment from affecting the correct data point after unification; Based on the unified correct data points, verify the data source in the external environment to check whether the data source in the external environment is consistent with the unified correct data points; When the correct data point after unification increases or decreases, it indicates that there is clinical inconsistency in children's respiratory infectious diseases and is marked as False; Otherwise, when the correct data points after unification are consistent, it means that there is clinical consistency in children's respiratory infectious diseases, which is marked as True. The specific steps of the verification are: When Records When it is equal to True, the verification passes; Otherwise, when Records If it is equal to False, the verification fails.

[0031] S2.1: Multi-source data fusion of reference data and distributed computing framework includes using the task scheduler of the distributed computing framework to allocate data processing tasks of the reference data to different distributed computing framework nodes; Design the synchronization mechanism of reference data according to the distributed computing framework nodes, keep the distributed computing framework nodes consistent with the correct data points after unification, and set up the exception handling synchronization mechanism; When the distributed computing framework nodes are consistent with the correct data points after unification, the multi-source data fusion is normal; When the distributed computing framework node is inconsistent with the correct data point after unification, the multi-source data fusion is abnormal. At this time, the administrator is notified to conduct an investigation.

[0032] S3: Construct a clinical intelligent diagnosis model based on the fused multi-source data, establish a dynamic monitoring and early warning system for the clinical intelligent diagnosis model through Internet technology, track changes in the health status of child patients, and complete clinical decision-making for children's respiratory infections.

[0033] Among them, building a clinical intelligent diagnosis model based on the fused multi-source data includes preprocessing the fused multi-source data, and the preprocessing includes feature selection and feature scaling of the multi-source data fusion; Feature selection includes selecting the diagnostic results for clinical decision-making of children's respiratory infections. The specific formula is: ; in, represents the selected feature set, The diagnostic result eigenvalues, represents the feature selection threshold; Feature scaling involves converting the first The eigenvalues ​​are scaled to a uniform range and unified using the normalization calculation method. The normalization calculation formula is: ; in, represents the scaled normalized eigenvalue, The diagnostic result The mean of the eigenvalues, The diagnostic result The standard deviation of the eigenvalues.

[0034] Furthermore, when performing feature selection, the feature selection threshold is used to screen features that have a significant impact on the diagnostic results. Specifically, the feature selection threshold is usually set to a value between 0 and 1 to accommodate different feature importance distributions. For example, if the feature selection threshold is set to 0.8, this means that only those features with an importance score greater than 0.8 will be selected into the final feature set. This ensures that the features used by the model have a significant impact on the diagnostic results, thereby improving the accuracy and robustness of the model. For example, if the importance score of a feature is 0.9, which is greater than the threshold of 0.8, the feature will be selected into the final feature set; conversely, if the importance score of a feature is 0.7, which is less than the threshold of 0.8, the feature will be excluded. The feature selection threshold data is shown in Table 1 below: Table 1 Feature selection threshold data table Parameter name Numeric describe Feature selection threshold 0.8 The importance threshold used to filter out features that have a significant impact on the diagnosis results Feature Importance Score 1 0.9 If the value is greater than the threshold value 0.8, the feature will be selected into the final feature set. Feature Importance Score 2 0.7 If the value is less than the threshold value 0.8, the feature will be excluded. Table 1 clearly shows the thresholds used in the feature selection process and the importance scores of specific features, as well as the relationship between these scores and the thresholds.

[0035] S3.1: Dynamic monitoring and early warning of the establishment of a clinical intelligent diagnosis model includes using the clinical intelligent diagnosis model to predict the scaled normalized feature values, tracking changes in the health status of child patients, and determining whether the child patients are at risk of infection; The prediction result calculation formula for predicting the scaled normalized eigenvalue is: ; in, Indicates at time The prediction results, Indicates time, Indicates the real time at this moment; when The prediction result of real time is greater than the feature selection threshold When the pediatric respiratory infection disease clinical decision support system issues an early warning, the clinical decision support system for pediatric respiratory infection disease is completed according to the set user authority; Otherwise, when The prediction result of real time is less than or equal to the feature selection threshold When the child is monitored, the changes in health status of the child patient can be tracked.

[0036] In a preferred embodiment, a clinical decision support system for children's respiratory infections is provided, which includes an application module, which uses the clinical decision support system for children's respiratory infections to mine a distributed computing framework for children's respiratory infections; a collection module, which collects clinical reference data of children's respiratory infections in an external environment and fuses the reference data with the distributed computing framework for multi-source data; a decision module, which constructs a clinical intelligent diagnosis model based on the fused multi-source data, establishes dynamic monitoring and early warning of the clinical intelligent diagnosis model through Internet technology, tracks changes in the health status of child patients, and completes clinical decisions on children's respiratory infections.

[0037] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0038] In summary, the present invention effectively improves the quality and processing efficiency of clinical data on children's respiratory infectious diseases by using a distributed computing framework and an intelligent diagnostic model. It removes erroneous data points through a data point cleaning algorithm to ensure the accuracy and consistency of the data. It integrates information from different data sources through multi-source data fusion technology to improve the comprehensiveness and reliability of the data. It timely tracks changes in the health status of child patients through real-time dynamic monitoring and early warning, and achieves early identification and precise treatment. In addition, the present invention also designs a strict user authority management and data verification mechanism to ensure data security and privacy protection, provide strong support for clinical decision-making, and significantly improve the diagnosis and treatment level of children's respiratory infectious diseases.

[0039] Example 2: Reference Figure 1 and Figure 2 , which is the second embodiment of the present invention, provides a clinical decision support system and method for children's respiratory infectious diseases. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0040] During the experiment, a clinical dataset of pediatric respiratory infectious diseases containing 10,000 records was used. Each record contained 30 features. There were 10,000 records in the initial dataset. After the data point cleaning algorithm removed the erroneous data points, 9,500 valid records remained. A total of 500 records containing erroneous data points were found and removed, and the number of missing feature values ​​was reduced by 5%. Then, the 9,500 valid records were standardized, and all feature values ​​were scaled to the range of 0 to 1. The mean of the feature value was 0.5 and the standard deviation was 0.2. The feature selection threshold of 0.75 was used to screen out features that had a significant impact on the diagnostic results. Finally, 20 features were selected into the model, accounting for 66.7% of the total number of features. There were 20 features with feature importance scores above 0.75, the highest feature importance score was 0.95, and the lowest was 0.76. The 20 features selected into the model were normalized to ensure that the eigenvalues ​​were in the same range. The normalized eigenvalue range was 0 to 1, the mean of the eigenvalues ​​was 0.48, and the standard deviation was 0.15. The random forest model was used for training, and the ratio of the training set to the test set was 70:30. The training set contained 6650 records, and the test set contained 2850 records. During the training process, the model had an accuracy of 92%, a recall rate of 88%, and an F1 score of 90%. On the test set, the model had an accuracy of 91%, a recall rate of 87%, and an F1 score of 89%. Finally, a dynamic monitoring and early warning system for the clinical intelligent diagnosis model was established through Internet technology, which tracked the changes in the health status of pediatric patients in real time, successfully identified and warned 150 potential infection risks, and the false alarm rate was only 2%. The experimental data of the present invention are shown in Table 2 below: Table 2 Experimental data table of the present invention Parameter name Numeric describe Initial total number of records 10000 The total number of records in the initial dataset Total number of valid records after cleaning 9500 The total number of valid records remaining after data cleaning Number of error records removed 500 Error record data removed during data cleaning Reduction ratio of missing feature values 5% The ratio of the number of missing values ​​of features reduced after data cleaning Eigenvalue mean 0.5 The mean of the eigenvalues ​​after standardization Eigenvalue standard deviation 0.2 Standard deviation of the eigenvalue after normalization Feature selection threshold 0.75 Threshold for feature selection The number of features finally selected into the model 20 The number of features selected into the model after feature selection Feature selection ratio 66.7% The ratio of the number of features selected into the model to the total number of features Highest feature importance score 0.95 The feature with the highest feature importance score Minimum feature importance score 0.76 The feature with the lowest feature importance score Normalized eigenvalue mean 0.48 The mean of the eigenvalues ​​after normalization Standard deviation of eigenvalues ​​after normalization 0.15 Standard deviation of the eigenvalue after normalization Number of training set records 6650 The number of records in the training set Number of test set records 2850 The number of records in the test set Training set accuracy 92% The accuracy of the model on the training set Training set recall 88% The recall rate of the model on the training set Training set F1 score 90% The F1 score of the model on the training set Test set accuracy 91% The accuracy of the model on the test set Test set recall 87% The recall rate of the model on the test set Test set F1 score 89% The F1 score of the model on the test set Successfully identified potential infection risks 150 cases Number of potential infection risk cases successfully identified by the dynamic monitoring and early warning system False Positive Rate 2% False alarm rate of dynamic monitoring and early warning system Table 2 clearly shows the data and description of each key step in the experimental process. The comparison between the present invention and the prior art is shown in Table 3 below: Table 3 Comparison table between the present invention and the prior art Compare Projects Prior art Technical solution of the present invention Advantage Analysis Data processing efficiency Single machine processing, slow processing speed, difficult to process large-scale data Distributed computing framework to efficiently process large-scale data Using a distributed computing framework to significantly improve data processing speed and efficiency Data quality The data cleaning is not thorough, and there are many erroneous data points. Data point cleaning algorithm to remove erroneous data points Ensure data accuracy and consistency through data point cleaning algorithms Feature Selection Manual feature selection can easily miss important features Automatic feature selection, selecting features that have a significant impact on the diagnosis results Automatically select important features to improve model accuracy and robustness Feature Scaling Lack of unified feature scaling method, inconsistent feature value ranges Normalization processing, unified eigenvalue range Unify the range of eigenvalues ​​to ensure the stability and accuracy of model training Model training and evaluation Model training time is long and evaluation indicators are average Fast training, high precision and high recall The model training speed is fast, the evaluation indicators are excellent, and the accuracy and reliability of diagnosis are improved Dynamic monitoring and early warning Lack of real-time monitoring and early warning functions Real-time dynamic monitoring and early warning, timely tracking of health status changes Realize real-time monitoring and early warning, timely detect potential infection risks, and improve early identification capabilities User Rights Management Lack of strict user rights management, low data security Set user permissions to ensure data security Strictly manage user permissions to ensure data security and privacy protection Exception handling mechanism Lack of effective exception handling mechanism affects data consistency Set up exception handling synchronization mechanism to ensure data consistency Ensure data consistency and integrity through exception handling mechanism Experimental Results Low precision and recall, high false positive rate Precision rate 91%, recall rate 87%, false alarm rate 2% The experimental results show that the present invention has higher accuracy, recall rate and lower false alarm rate. Table 3 clearly shows the significant advantages of the present invention in data processing efficiency, data quality, feature selection, feature scaling, model training and evaluation, dynamic monitoring and early warning, user authority management, and exception handling mechanism. These advantages enable the present invention to be better applied to the clinical decision support system for children's respiratory infections and improve the accuracy and reliability of diagnosis.

[0041] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A clinical decision support method for children's respiratory tract infections, characterized by: include: Using the clinical decision support system for children's respiratory infections, we explored the distributed computing framework for children's respiratory infections; Collect clinical reference data of pediatric respiratory infectious diseases in the external environment, and fuse the reference data with the distributed computing framework for multi-source data; A clinical intelligent diagnosis model is constructed based on the fused multi-source data, and a dynamic monitoring and early warning system for the clinical intelligent diagnosis model is established through Internet technology to track changes in the health status of child patients and complete clinical decision-making for children's respiratory infections.

2. The clinical decision support method for children's respiratory tract infectious diseases according to claim 1, characterized in that: The use of the clinical decision support system for children's respiratory tract infection diseases includes using a data point cleaning algorithm to remove erroneous data points of clinical symptoms, and unifying the data format of correct data points of clinical symptoms through standardization processing, wherein the use of the data point cleaning algorithm includes calculating the number of missing values ​​of clinical characteristics of children's respiratory tract infection diseases, calculating the proportion of the number of missing values ​​of clinical characteristics of children's respiratory tract infection diseases according to the number of missing values ​​of characteristics, and calculating the average proportion of the number of missing values ​​of characteristics based on the proportion of the number of missing values ​​of characteristics; The formula for calculating the number of missing values ​​of clinical characteristics of children's respiratory infections is: ； in, Indicates Record No. eigenvalues, represents the indicator function, Indicates a record bar. Indicates the number of missing values ​​of features; When Record No. When the eigenvalue returns 1, there is a missing value; Otherwise, when Record No. When the feature value returns 0, there is no missing value; The formula for the proportion of missing values ​​of the feature is: ； in, represents the proportion of missing values ​​of features, represents the number of missing values ​​of features, Indicates a record bar.

3. The clinical decision support method for children's respiratory tract infection diseases according to claim 2, characterized in that: The formula for the average proportion of the number of missing values ​​of the feature is: ； in, represents the average proportion of missing values ​​of features, Indicates the number of features of the record bar, Indicates the proportion of missing values ​​of features; The distributed computing framework for mining children's respiratory tract infection diseases includes calculating the distribution standard deviation of the number of feature missing values ​​according to the average proportion of the number of feature missing values, and the formula for the distribution standard deviation is: ； in, represents the standard deviation of the distribution; The threshold of the distributed computing framework is calculated according to the distribution standard deviation, and the formula of the threshold is: ； in, represents the adjustment factor, represents the threshold of the distributed computing framework; When the number of missing values ​​of features is proportional Greater than threshold When , the number of missing feature values ​​is removed. At this time, the number of missing feature values ​​is the wrong data point of clinical symptoms; When the number of missing values ​​of features is proportional Less than or equal to threshold When , it is the correct data point of the clinical symptoms. At this time, the data format is unified by filling the mean, median and interpolation method of the proportion of the number of missing values ​​of the features.

4. The clinical decision support method for children's respiratory tract infection diseases according to claim 3, characterized in that: The collecting of clinical reference data of respiratory infectious diseases in children under the external environment includes processing the data source under the external environment and the unified correct data point, and the processing includes setting user permissions and isolating the data source under the external environment from the unified correct data point; The setting of user permissions includes allowing staff to access unified pediatric patient data at correct data points; Otherwise, when user permissions are not set, only ordinary users are allowed to view the summary data; The isolating the data source in the external environment from the unified correct data point includes preventing the data source in the external environment from affecting the unified correct data point; Based on the unified correct data points, verify the data source in the external environment to check whether the data source in the external environment is consistent with the unified correct data points; When the correct data point after unification increases or decreases, it indicates that there is clinical inconsistency in children's respiratory infectious diseases and is marked as False; Otherwise, when the correct data points after unification are consistent, it means that there is clinical consistency in children's respiratory infectious diseases, which is marked as True. The specific steps of the verification are: When Records When it is equal to True, the verification passes; Otherwise, when Records If it is equal to False, the verification fails.

5. The clinical decision support method for children's respiratory tract infection diseases according to claim 4, characterized in that: The multi-source data fusion of the reference data and the distributed computing framework includes using a task scheduler of the distributed computing framework to distribute data processing tasks of the reference data to different distributed computing framework nodes; Design the synchronization mechanism of reference data according to the distributed computing framework nodes, keep the distributed computing framework nodes consistent with the correct data points after unification, and set up the exception handling synchronization mechanism; When the distributed computing framework nodes are consistent with the correct data points after unification, the multi-source data fusion is normal; When the distributed computing framework node is inconsistent with the correct data point after unification, the multi-source data fusion is abnormal. At this time, the administrator is notified to conduct an investigation.

6. The clinical decision support method for children's respiratory tract infection diseases according to claim 5, characterized in that: The constructing of the clinical intelligent diagnosis model based on the fused multi-source data includes preprocessing the fused multi-source data, wherein the preprocessing includes feature selection and feature scaling of the multi-source data fusion; The feature selection includes selecting the diagnostic results of clinical decision-making for children's respiratory tract infectious diseases, and the specific formula is: ； in, represents the selected feature set, The diagnostic result eigenvalues, represents the feature selection threshold; The feature scaling includes scaling the first The eigenvalues ​​are scaled to a uniform range and unified using a normalized calculation method. The normalized calculation formula is: ； in, represents the scaled normalized eigenvalue, The diagnostic result The mean of the eigenvalues, The diagnostic result The standard deviation of the eigenvalues.

7. The clinical decision support method for children's respiratory tract infection diseases according to claim 6, characterized in that: The dynamic monitoring and early warning of establishing a clinical intelligent diagnosis model includes using the clinical intelligent diagnosis model to predict the scaled normalized eigenvalues, tracking the changes in the health status of the child patient, and determining whether the child patient has an infection risk; The prediction result calculation formula for predicting the scaled normalized eigenvalue is: ； in, Indicates at time The prediction results, Indicates time, Indicates the real time at this moment; when The prediction result of real time is greater than the feature selection threshold When the pediatric respiratory infection disease clinical decision support system issues an early warning, the clinical decision support system for pediatric respiratory infection disease is completed according to the set user authority; Otherwise, when The prediction result of real time is less than or equal to the feature selection threshold When the child is monitored, the changes in health status of the child patient can be tracked.

8. A clinical decision support system for children's respiratory tract infections, based on the clinical decision support method for children's respiratory tract infections according to any one of claims 1 to 7, characterized in that: include: The application module uses the clinical decision support system for children's respiratory infections to explore the distributed computing framework for children's respiratory infections; The acquisition module collects clinical reference data of children's respiratory infectious diseases in the external environment and integrates the reference data with the distributed computing framework for multi-source data fusion; The decision-making module builds a clinical intelligent diagnosis model based on the integrated multi-source data, establishes dynamic monitoring and early warning of the clinical intelligent diagnosis model through Internet technology, tracks changes in the health status of child patients, and completes clinical decision-making for children's respiratory infections.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the clinical decision support method for children's respiratory infectious diseases described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the clinical decision support method for children's respiratory infectious diseases described in any one of claims 1 to 7 are implemented.

Images