Bronchial lung dysplasia severity prediction method and related equipment

Abstract

The invention discloses a bronchial lung dysplasia severity prediction method and related equipment, and belongs to the technical field of machine learning, and the method comprises the steps: obtaining the clinical data of a target object; preprocessing the clinical data of the target object, constructing a training set and a verification set according to a preset proportion, and screening to obtain a target variable; based on the target variable, constructing a plurality of bronchial lung dysplasia severity prediction models by using the training set, and performing model performance evaluation on the plurality of constructed bronchial lung dysplasia severity prediction models in the verification set; and obtaining a bronchial lung dysplasia severity prediction model with the optimal performance so as to complete bronchial lung dysplasia severity prediction. The clinical data of the target object is combined with the machine learning algorithm, so that the dynamic change of the immune system of the premature infant can be reflected more comprehensively, the prediction accuracy of the severity of early bronchial pulmonary dysplasia is improved, and scientific data support is provided for clinical treatment selection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of machine learning, and in particular to a bronchopulmonary dysplasia severity prediction method and related equipment. BACKGROUND

[0002] Currently, bronchopulmonary dysplasia (BPD) is one of the most common and most serious complications of premature infants, which is manifested as alveolar development stagnation and abnormal pulmonary angiogenesis. The occurrence of BPD is affected by multiple factors such as infection, hyperoxia exposure and mechanical ventilation, and the core pathological mechanism is closely related to the imbalance of immune inflammatory response.

[0003] In the related art, the existing BPD early prediction methods mostly depend on clinical data at birth, omics data or laboratory indicators at a single time point. However, in actual application, it is found that the traditional BPD prediction method is difficult to reflect the dynamic changes of the immune system of premature infants, and there are problems such as limited prediction accuracy.

[0004] In summary, the technical problems existing in the related art need to be improved. SUMMARY

[0005] The embodiments of the present application provide a bronchopulmonary dysplasia severity prediction method and related equipment, which can more comprehensively reflect the dynamic changes of the immune system of premature infants, improve the prediction accuracy of the early bronchopulmonary dysplasia severity, and thus provide scientific data support for clinical treatment selection.

[0006] In one aspect, the embodiments of the present application provide a bronchopulmonary dysplasia severity prediction method, which comprises the following steps: obtaining clinical data of a target object; preprocessing the clinical data of the target object, dividing a training set and a validation set according to a preset proportion, and screening and obtaining target variables; based on the target variables, constructing a plurality of bronchopulmonary dysplasia severity prediction models using the training set, and performing model performance evaluation on a plurality of bronchopulmonary dysplasia severity prediction models that have completed training according to the validation set; wherein the model construction basis of the plurality of bronchopulmonary dysplasia severity prediction models are all different; obtaining a bronchopulmonary dysplasia severity prediction model with optimal model performance evaluation result, for completing bronchopulmonary dysplasia severity prediction.

[0007] Optionally, based on the target variables, constructing a plurality of bronchopulmonary dysplasia severity prediction models using the training set, and performing model performance evaluation on a plurality of bronchopulmonary dysplasia severity prediction models that have completed training according to the validation set, comprises: extracting feature variables within a first preset time point, feature variables within a second preset time point, feature variables within a third preset time point, and feature variables within a fourth preset time point according to the training set based on the target variable; constructing multiple bronchopulmonary dysplasia severity prediction models based on different machine learning algorithms as a model construction basis based on the feature variables within the first preset time point; constructing multiple bronchopulmonary dysplasia severity prediction models based on different machine learning algorithms as a model construction basis based on the feature variables within the second preset time point; constructing multiple bronchopulmonary dysplasia severity prediction models based on different machine learning algorithms as a model construction basis based on the feature variables within the third preset time point; constructing multiple bronchopulmonary dysplasia severity prediction models based on different machine learning algorithms as a model construction basis based on the feature variables within the fourth preset time point; performing model performance evaluation on all bronchopulmonary dysplasia severity prediction models at all preset time points according to the validation set; The different machine learning algorithms include random forest, artificial neural network, naive Bayes, and regression tree.

[0008] Optionally, the model performance evaluation on all bronchopulmonary dysplasia severity prediction models at all preset time points according to the validation set comprises: evaluating all bronchopulmonary dysplasia severity prediction models through multiple receiver operating characteristic curves according to the validation set; determining the prediction performance of each prediction model on bronchopulmonary dysplasia severity through the area under the receiver operating characteristic curve.

[0009] Optionally, the preprocessing of the target object clinical data, the division of the training set and the validation set according to the preset proportion, and the screening and obtaining of the target variable comprise: using the K-nearest neighbor method to fill in the missing values in the target object clinical data; dividing the data into the training set and the validation set according to the preset proportion; screening variables related to bronchopulmonary dysplasia in the target object clinical data through the least absolute selection and shrinkage operator algorithm, the random forest algorithm, and the support vector machine recursive feature elimination algorithm respectively; taking the intersection of the screening results based on the least absolute selection and shrinkage operator algorithm, the random forest algorithm, and the support vector machine recursive feature elimination algorithm to obtain the target variable.

[0010] Optionally, the obtaining of the target object clinical data comprises: obtain blood routine index data and clinical variables of each target object as target object clinical data; The blood routine index data includes white blood cell count, neutrophil count, lymphocyte count, and platelet count, and the clinical variables include gestational age, birth weight, and gender.

[0011] Optionally, after the target object clinical data is obtained, the method further includes: According to a preset diagnosis standard, the target object is clinically diagnosed and graded for bronchopulmonary dysplasia, and the diagnosis grading result includes no bronchopulmonary dysplasia, mild bronchopulmonary dysplasia, moderate bronchopulmonary dysplasia, and severe bronchopulmonary dysplasia.

[0012] In another aspect, an embodiment of the present application provides a bronchopulmonary dysplasia severity prediction device, the device comprising: a data acquisition module configured to acquire target object clinical data; a data processing module configured to preprocess the target object clinical data, divide a training set and a validation set according to a preset ratio, and filter and obtain target variables; a training and verification module configured to construct a plurality of bronchopulmonary dysplasia severity prediction models based on the target variables using the training set, and perform model performance evaluation on the plurality of bronchopulmonary dysplasia severity prediction models that have completed training according to the validation set; wherein the plurality of bronchopulmonary dysplasia severity prediction models are based on different model construction bases; a model prediction module configured to obtain a bronchopulmonary dysplasia severity prediction model with optimal model performance evaluation result, to be used for completing bronchopulmonary dysplasia severity prediction.

[0013] In another aspect, an embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0014] In another aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above method.

[0015] In another aspect, an embodiment of the present application provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the above method.

[0016] The embodiment of the application can more comprehensively reflect the dynamic changes of the immune system of premature infants by combining the target object clinical data with the machine learning algorithm, improve the early prediction accuracy of the severity of bronchopulmonary dysplasia, and thereby provide scientific data support for clinical treatment selection. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is an implementation environment schematic diagram of a bronchopulmonary dysplasia severity prediction method provided by the embodiment of the application; Figure 2 is a flow schematic diagram of a bronchopulmonary dysplasia severity prediction method provided by the embodiment of the application; Figure 3 is a schematic diagram of screening and obtaining a target variable provided by the embodiment of the application; Figure 4 is a schematic diagram of a receiver operating characteristic curve of a bronchopulmonary dysplasia severity prediction model provided by the embodiment of the application; Figure 5 is a structural schematic diagram of a bronchopulmonary dysplasia severity prediction device provided by the embodiment of the application; Figure 6 is a hardware structure schematic diagram of an electronic device provided by the embodiment of the application. DETAILED DESCRIPTION

[0018] In order to make the objects, technical solutions and advantages of the application clearer, the application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and not to limit the application. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementation modes described in the following exemplary embodiments do not represent all the implementation modes consistent with the embodiments of the application, but are only examples of devices and methods consistent with some aspects of the embodiments of the application as detailed in the appended claims.

[0019] It can be understood that the terms “first”, “second”, and the like used in the application can be used herein to describe various concepts, but unless specifically stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the application, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information. Depending on the context, the word “if” as used herein can be interpreted as “when” or “when” or “in response to determining”.

[0020] As used herein, the terms “at least one”, “multiple”, “each”, “any”, and the like, at least one includes one, two, or more, multiple includes two or more, each refers to each of the corresponding plurality, and any refers to any one of the plurality.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to be limiting of this application.

[0022] Currently, bronchopulmonary dysplasia (BPD) is one of the most common and severe complications of premature infants, which is manifested as arrested alveolar development and abnormal pulmonary angiogenesis. The occurrence of BPD is affected by multiple factors such as infection, hyperoxia exposure and mechanical ventilation, and its core pathological mechanism is closely related to the imbalance of immune inflammatory response.

[0023] In the related art, most of the existing early prediction methods of BPD rely on clinical data at birth, omics data or laboratory indicators at a single time point. However, it is found in actual application that the traditional BPD prediction method is difficult to reflect the dynamic changes of the immune system of premature infants, and has problems such as limited prediction accuracy.

[0024] Therefore, in the embodiments of the present application, a bronchopulmonary dysplasia severity prediction method and related equipment are provided, which can more comprehensively reflect the dynamic changes of the immune system of premature infants by combining the clinical data of the target object with the machine learning algorithm, improve the prediction accuracy of the early bronchopulmonary dysplasia severity, and thus provide scientific data support for clinical treatment selection.

[0025] It should be noted that in each specific embodiment of the present application, when it is necessary to perform relevant processing according to user information, user behavior data, user historical data, and user location information and other data related to the identity or characteristics of the user, the user's permission or consent will be obtained first, and the collection, use and processing of these data will comply with relevant laws, regulations and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or jumping to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for enabling the embodiments of the present application to normally operate will be obtained.

[0026] The specific embodiments of the embodiments of the present application will be described in detail below with reference to the accompanying drawings. First, a bronchopulmonary dysplasia severity prediction method provided in the embodiments of the present application is described with reference to the accompanying drawings.

[0027] Please refer to Figure 1 ,Figure 1 This is a schematic diagram of the implementation environment for a method for predicting the severity of bronchopulmonary dysplasia provided in this application embodiment. In this implementation environment, the main hardware and software components involved include a terminal processor 110 and a server 120.

[0028] Specifically, the terminal processor 110 may be equipped with a control program for predicting the severity of bronchopulmonary dysplasia, and the server 120 serves as the backend server for this control program. The terminal processor 110 and the backend server 120 are connected in communication. The bronchopulmonary dysplasia severity prediction method provided in this embodiment can be executed on the terminal processor 110 side.

[0029] Server 120 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0030] In addition, server 120 can also be a node server in a blockchain network.

[0031] The terminal processor 110 and the server 120 can establish a communication connection via a wireless network. This wireless network uses standard communication technologies and / or protocols. The network can be the Internet or any other network, including but not limited to a Local Area Network (LAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), mobile, or any combination of wireless networks, private networks, or virtual private networks. Furthermore, these hardware and software components can use the same or different communication connection methods; this application does not impose specific limitations in this regard.

[0032] Of course, this is understandable. Figure 1 The implementation environment described in this application is only one of the optional application scenarios for the bronchopulmonary dysplasia severity prediction method provided in this embodiment. The actual application is not fixed. Figure 1 The software and hardware environment shown is not specifically limited in this application.

[0033] like Figure 2 As shown, Figure 2This is a flowchart illustrating a method for predicting the severity of bronchopulmonary dysplasia provided in an embodiment of this application, specifically including but not limited to steps 100 to 400.

[0034] Step 100: Obtain clinical data of the target subjects.

[0035] In this application embodiment, bronchopulmonary dysplasia (BPD) is usually the most common and most serious complication of premature infants, characterized by alveolar development arrest and abnormal pulmonary angiogenesis. Therefore, the target subjects can be premature infants in various hospitals, and clinical data of the target subjects can be obtained by accessing the data system of any hospital.

[0036] For example, by accessing the medical record system of any hospital, structured and semi-structured data from electronic medical records can be read to obtain data from obstetrics or neonatology as clinical data for the target.

[0037] Specifically, as an optional implementation, acquiring the clinical data of the target object includes: Obtain the blood routine index data and clinical variables of each target object as the target object's clinical data; The blood routine indicators include white blood cell, neutrophil, lymphocyte, and platelet counts; the clinical variables include gestational age, birth weight, and sex.

[0038] In the embodiments of this application, each target object (e.g., each premature infant) can be considered as a unit by obtaining the complete blood count (CBC) data and clinical variables of each target object. The CBC data mainly includes the number of white blood cells, neutrophils, lymphocytes, and platelets, while the clinical variables include gestational age, birth weight, and sex.

[0039] For example, by collecting medical records of preterm infants with a gestational age of <32 weeks, the CBC index of each preterm infant at birth, on the 2nd, 7th, 14th, 21st and 28th day after birth was extracted, and the gestational age, birth weight and sex of each preterm infant were read, thus serving as clinical data for the target subjects.

[0040] Furthermore, after obtaining the clinical data of the target subjects, the clinical diagnosis of BPD can be graded for each target subject based on the preset diagnostic criteria, and can be divided into asymptomatic bronchopulmonary dysplasia, mild bronchopulmonary dysplasia, moderate bronchopulmonary dysplasia, and severe bronchopulmonary dysplasia according to the severity.

[0041] For example, when grading BPD diagnosis based on clinical data for each target subject, the presence of BPD can first be determined by whether postnatal oxygen dependence (≥21%) exceeds 28 days. That is, postnatal oxygen dependence less than 28 days is diagnosed as no BPD, and postnatal oxygen dependence greater than or equal to 28 days is diagnosed as having BPD. Furthermore, BPD severity can be categorized into mild, moderate, and severe BPD. Specifically, when the corrected gestational age (i.e., gestational age + postnatal age) reaches 36 weeks, mild BPD is diagnosed without oxygen support; when the corrected gestational age reaches 36 weeks, moderate BPD is diagnosed with <30% oxygen support but no mechanical ventilation; and when the corrected gestational age reaches 36 weeks, severe BPD is diagnosed with ≥30% oxygen support or continuous positive airway pressure. Moderate and severe BPD require long-term oxygen support, severely impacting the infant's health and quality of life, and have the worst prognosis.

[0042] Therefore, by classifying each target object for BPD clinical diagnosis, it can be used as a sample label for subsequent machine learning prediction model construction.

[0043] Step 200: Preprocess the clinical data of the target object, divide the training set and validation set according to a preset ratio, and select and obtain the target variables.

[0044] In the embodiments of this application, the acquired clinical data of the target object can be preprocessed. For example, data cleaning can be performed first, duplicate data records can be deleted, units can be standardized, and CBC index data at multiple time points can be time-series aligned. Furthermore, missing values ​​can be filled in.

[0045] Specifically, as an optional implementation, the preprocessing of the clinical data of the target object, dividing the training set and validation set according to a preset ratio, and selecting and obtaining the target variables includes: The K-nearest neighbor method was used to impute missing values ​​in the clinical data of the target subjects; The data is divided into training and validation sets according to a preset ratio. The variables related to bronchopulmonary dysplasia in the clinical data of the target object were screened by the minimum absolute value selection and contraction operator algorithm, the random forest algorithm, and the support vector machine recursive feature elimination algorithm, respectively. The target variable is obtained by taking the intersection of the filtering results from the minimum absolute value selection and shrinkage operator algorithm, the random forest algorithm, and the support vector machine recursive feature elimination algorithm.

[0046] In this embodiment of the application, when preprocessing the clinical data of the target object, the missing values ​​in the clinical data of the target object can be filled by using the K nearest neighbor method, and redundant features can be further removed to reduce the interference of noise and prevent the subsequent model from overfitting. Specifically, the variables related to BPD in the clinical data of the target object can be screened by using algorithms such as minimum absolute value selection and shrinkage operator, random forest, and support vector machine recursive feature elimination.

[0047] For example, please refer to Figure 3 , Figure 3 This is a schematic diagram of a method for screening and obtaining target variables provided in an embodiment of this application. After screening the variables related to BPD in the clinical data of the target object by using the minimum absolute value selection and shrinkage operator algorithm, the random forest algorithm and the support vector machine recursive feature elimination algorithm respectively, the intersection of the screening results of the three algorithms can be taken to determine the target variables that are highly related to BPD.

[0048] Furthermore, the data is randomly divided into training and validation sets at a preset ratio (e.g., 7:3), so that the selected target variable can be used as the input variable and the BPD diagnosis and grading results can be used as the outcome variable, and multiple BPD prediction models can be trained using the training set.

[0049] Step 300: Based on the target variable, construct multiple bronchopulmonary dysplasia severity prediction models using the training set, and evaluate the model performance of the multiple trained bronchopulmonary dysplasia severity prediction models according to the validation set. The underlying models for predicting the severity of bronchopulmonary dysplasia are all different.

[0050] In the embodiments of this application, multiple BPD prediction models can be constructed by applying different machine learning algorithms in the training set, and the performance of the trained BPD prediction models can be evaluated using the validation set, so as to select the BPD prediction model with the best prediction performance.

[0051] In practical applications, based on the target variable, multiple bronchopulmonary dysplasia severity prediction models are constructed using the training set, and the performance of these models is evaluated using the validation set, including: Based on the target variable, feature variables within the first preset time point, the second preset time point, the third preset time point, and the fourth preset time point are extracted according to the training set. Based on the feature variables within the first preset time point, multiple bronchopulmonary dysplasia severity prediction models are constructed using different machine learning algorithms as the model building basis. Based on the feature variables within the second preset time point, multiple bronchopulmonary dysplasia severity prediction models are constructed using different machine learning algorithms as the model building basis. Based on the feature variables within the third preset time point, multiple bronchopulmonary dysplasia severity prediction models are constructed using different machine learning algorithms as the model building basis. Based on the feature variables within the fourth preset time point, multiple bronchopulmonary dysplasia severity prediction models are constructed using different machine learning algorithms as the model building basis. Based on the validation set, the performance of all bronchopulmonary dysplasia severity prediction models at all preset time points was evaluated. The different machine learning algorithms mentioned include random forest, artificial neural network, Naive Bayes, and regression tree.

[0052] In this embodiment of the application, in the constructed training set, feature variables within a first preset time point, feature variables within a second preset time point, feature variables within a third preset time point, and feature variables within a fourth preset time point are extracted according to the time points corresponding to the feature variables.

[0053] For example, the first preset time point can be set to within 1 week after birth, the second preset time point to within 2 weeks after birth, the third preset time point to within 3 weeks after birth, and the fourth preset time point to within 4 weeks after birth, thereby dividing the feature variables in the training set into multiple consecutive time periods.

[0054] Furthermore, for each preset time point, the feature variables are used to construct multiple bronchopulmonary dysplasia severity prediction models based on different machine learning algorithms.

[0055] For example, taking CBC index data as the target variable, for CBC index data within 1 week after birth, four BPD prediction models are constructed based on random forest, artificial neural network, naive Bayes, and regression tree respectively; for CBC index data within 2 weeks after birth, four BPD prediction models are constructed based on random forest, artificial neural network, naive Bayes, and regression tree respectively, and so on, until the BPD prediction models for the four preset time points are constructed.

[0056] Furthermore, the performance of all bronchopulmonary dysplasia severity prediction models at all preset time points was evaluated using a validation set, thereby selecting the best-performing BPD prediction model.

[0057] In practical applications, all bronchopulmonary dysplasia severity prediction models can be evaluated using multiple receiver operating characteristic (ROC) curves, and then the predictive performance of each prediction model for bronchopulmonary dysplasia severity can be determined by the area under the curve (AUC) of the multiple ROC curves.

[0058] For example, please refer to Figure 4 , Figure 4 This is a schematic diagram of the receiver operating characteristic curve of a bronchopulmonary dysplasia severity prediction model provided in an embodiment of this application. Figure 4 The figure shows the ROC curves of a random forest model constructed using CBC indicators within one week after birth as the training set. Specificity is plotted on the horizontal axis, sensitivity on the vertical axis, and the AUC for different BPD prediction results is indicated. Specifically, the gray line in the ROC curve is the baseline; the blue ROC curve corresponds to a diagnosis of no BPD, the yellow ROC curve to a diagnosis of mild BPD, and the red curve to a diagnosis of moderate to severe BPD. The AUC for each ROC curve is calculated: 0.8878 for the blue ROC curve, 0.6901 for the yellow ROC curve, and 0.9354 for the red ROC curve.

[0059] Among them, the AUC value specifically reflects the predictive ability of each type of BPD (the closer the value is to 1, the better the predictive ability). Therefore, according to Figure 4 It can be seen that the BPD prediction model constructed by combining CBC index data from multiple consecutive time points with the random forest algorithm model is very effective in identifying BPD without and moderate to severe BPD, and has high prediction accuracy.

[0060] Furthermore, by comparing the ROC curves of the regression tree model, random forest model, Naive Bayes model, and artificial neural network model, it can be determined that the random forest model has the highest overall accuracy of 0.761, and the confidence interval can be further calculated.

[0061] In practical applications, by comparing the ROC curves of BPD prediction models built with the same machine learning algorithm at different preset time points, it can be determined that the prediction performance of a BPD prediction model built using CBC index data from more consecutive time points is better.

[0062] Therefore, compared with the existing BPD prediction methods based on clinical variables, omics (such as transcriptomics and metabolomics) or single time point laboratory indicators, which have low accuracy, this application can combine continuous blood routine data with machine learning algorithms to predict the severity of BPD in the early clinical stage. Compared with traditional single time point data, it can more comprehensively reflect the dynamic changes of the immune system in preterm infants and guide clinical treatment selection.

[0063] Step 400: Obtain the bronchopulmonary dysplasia severity prediction model with the best model performance evaluation results, and use it to complete the prediction of bronchopulmonary dysplasia severity.

[0064] In this embodiment, by comparing the ROC curves of BPD prediction models built on different machine learning algorithms, the BPD prediction model with the best model performance evaluation result can be selected, and then the BPD prediction can be completed using the BPD prediction model with the best model performance evaluation result.

[0065] In practical applications, for example, a clinical application mini-program can be developed based on the BPD prediction model with the best model performance evaluation results. By inputting the clinical variables of premature infants and CBC indicators, it is possible to predict whether BPD will occur and its severity, thereby providing scientific data support for clinical treatment selection.

[0066] Please see Figure 5 , Figure 5 This is a schematic diagram of a bronchopulmonary dysplasia severity prediction device provided in an embodiment of this application. This application also provides a bronchopulmonary dysplasia severity prediction device that can implement the above-mentioned bronchopulmonary dysplasia severity prediction method. The device includes: Data acquisition module 510 is used to acquire clinical data of the target object; The data processing module 520 is used to preprocess the clinical data of the target object and filter to obtain target variables, thereby constructing a training set and a validation set based on the target variables and according to a preset ratio. The training and validation module 530 is used to construct multiple bronchopulmonary dysplasia severity prediction models based on the training set, and to evaluate the model performance of the multiple trained bronchopulmonary dysplasia severity prediction models based on the validation set; wherein the multiple bronchopulmonary dysplasia severity prediction models are all based on different model construction foundations. The model prediction module 540 is used to obtain the bronchopulmonary dysplasia severity prediction model with the best model performance evaluation results, so as to complete the prediction of the severity of bronchopulmonary dysplasia.

[0067] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0068] Please see Figure 6 , Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. The electronic device includes: The processor 601 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 602 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 602 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 602 and is called and executed by the processor 601 using the methods described in the embodiments of this application. The input / output interface 603 is used to implement information input and output; The communication interface 604 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 605 transmits information between various components of the device (e.g., processor 601, memory 602, input / output interface 603, and communication interface 604); The processor 601, memory 602, input / output interface 603, and communication interface 604 are connected to each other within the device via bus 605.

[0069] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0070] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0071] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0072] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0073] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0074] This application provides a method and related equipment for predicting the severity of bronchopulmonary dysplasia. By combining clinical data of the target subject with machine learning algorithms, it can more comprehensively reflect the dynamic changes of the immune system of premature infants, improve the accuracy of predicting the severity of early bronchopulmonary dysplasia, and thus provide scientific data support for clinical treatment selection.

[0075] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0076] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0077] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0078] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0079] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0080] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0081] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0082] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0083] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0084] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method of predicting the severity of bronchopulmonary dysplasia, characterized by, The method comprises the following steps: Obtaining target object clinical data; Pretreating the target object clinical data, dividing a training set and a verification set according to a preset proportion, and screening target variables; Based on the target variables, a plurality of bronchopulmonary dysplasia severity prediction models are constructed using the training set, and the model performance of the plurality of trained bronchopulmonary dysplasia severity prediction models is evaluated according to the verification set; wherein the model construction basis of the plurality of bronchopulmonary dysplasia severity prediction models is different; Obtaining the bronchopulmonary dysplasia severity prediction model with the optimal model performance evaluation result, for completing the bronchopulmonary dysplasia severity prediction.

2. The method of claim 1, wherein, Based on the target variables, a plurality of bronchopulmonary dysplasia severity prediction models are constructed using the training set, and the model performance of the plurality of trained bronchopulmonary dysplasia severity prediction models is evaluated according to the verification set, comprising: Based on the target variables, the feature variables within the first preset time point, the feature variables within the second preset time point, the feature variables within the third preset time point, and the feature variables within the fourth preset time point are extracted according to the training set; Based on the feature variables within the first preset time point, a plurality of bronchopulmonary dysplasia severity prediction models are constructed with different machine learning algorithms as the model construction basis; Based on the feature variables within the second preset time point, a plurality of bronchopulmonary dysplasia severity prediction models are constructed with different machine learning algorithms as the model construction basis; Based on the feature variables within the third preset time point, a plurality of bronchopulmonary dysplasia severity prediction models are constructed with different machine learning algorithms as the model construction basis; Based on the feature variables within the fourth preset time point, a plurality of bronchopulmonary dysplasia severity prediction models are constructed with different machine learning algorithms as the model construction basis; According to the verification set, the model performance of all bronchopulmonary dysplasia severity prediction models within all preset time points is evaluated; The different machine learning algorithms include random forest, artificial neural network, naive Bayes, and regression tree.

3. The method of claim 2, wherein, According to the verification set, the model performance of all bronchopulmonary dysplasia severity prediction models within all preset time points is evaluated, comprising: According to the verification set, all bronchopulmonary dysplasia severity prediction models are evaluated by the receiver operating characteristic curve; The prediction performance of each prediction model for bronchopulmonary dysplasia severity is determined by the area under the multiple receiver operating characteristic curve.

4. The method of claim 1, wherein, The pretreatment of the target object clinical data, the division of the training set and the verification set according to the preset proportion, and the screening of the target variables, comprising: The missing values in the target object clinical data are filled using the K-nearest neighbor method; The data is divided into a training set and a verification set according to a preset proportion; The variables related to bronchopulmonary dysplasia in the target object clinical data are screened by the least absolute selection and shrinkage operator algorithm, the random forest algorithm, and the support vector machine recursive feature elimination algorithm, respectively; An intersection of screening results based on a least absolute selection and shrinkage operator algorithm, a random forest algorithm, and a support vector machine recursive feature elimination algorithm is obtained to obtain the target variable.

5. The method of claim 1, wherein, The clinical data of the target object is obtained, including: Blood routine index data and clinical variables of each target object are obtained as the clinical data of the target object. The blood routine index data includes white blood cells, neutrophils, lymphocytes, and platelet counts, and the clinical variables include gestational age, birth weight, and gender.

6. The method of claim 1, wherein, After the clinical data of the target object is obtained, the method further includes: According to a preset diagnosis standard, each target object is clinically diagnosed and graded for bronchopulmonary dysplasia, and the diagnosis grading result includes no bronchopulmonary dysplasia, mild bronchopulmonary dysplasia, moderate bronchopulmonary dysplasia, and severe bronchopulmonary dysplasia.

7. A bronchopulmonary dysplasia severity prediction apparatus characterized by, The device includes: A data acquisition module is configured to obtain clinical data of a target object. A data processing module is configured to preprocess the clinical data of the target object, construct a training set and a validation set according to a preset proportion, and screen and obtain a target variable. A training and verification module is configured to construct a plurality of bronchopulmonary dysplasia severity prediction models based on the target variable using the training set, and perform model performance evaluation on the plurality of bronchopulmonary dysplasia severity prediction models that have completed training according to the validation set, wherein the model construction basis of the plurality of bronchopulmonary dysplasia severity prediction models is different. A model prediction module is configured to obtain a bronchopulmonary dysplasia severity prediction model with optimal model performance evaluation result for completing bronchopulmonary dysplasia severity prediction.

8. An electronic device, comprising: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method of any one of claims 1-6 when executing the computer program.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program is executed by the processor to implement the method of any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-6. The computer program is executed by the processor to implement the method of any one of claims 1-6.

Images