CN115101217A - Kawasaki disease aspirin resistance prediction model and prediction evaluation system - Google Patents
Kawasaki disease aspirin resistance prediction model and prediction evaluation system Download PDFInfo
- Publication number
- CN115101217A CN115101217A CN202210359448.7A CN202210359448A CN115101217A CN 115101217 A CN115101217 A CN 115101217A CN 202210359448 A CN202210359448 A CN 202210359448A CN 115101217 A CN115101217 A CN 115101217A
- Authority
- CN
- China
- Prior art keywords
- prediction
- kawasaki disease
- model
- aspirin resistance
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- Chemical & Material Sciences (AREA)
- Biomedical Technology (AREA)
- Medicinal Chemistry (AREA)
- Pharmacology & Pharmacy (AREA)
- Toxicology (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a Kawasaki disease aspirin resistance prediction model, which is characterized in that a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model are established by data input, the correctness, sensitivity, specificity, AUC, positive predicted values and negative predicted values of four digital prediction models are analyzed by adopting a confusion matrix and an ROC curve, variable weight processing is carried out on the prediction results of the four digital prediction models based on the analysis results, and the Kawasaki disease aspirin resistance prediction model is obtained; the invention provides a prediction evaluation system of a Kawasaki disease aspirin resistance prediction model, which is used for calculating Kawasaki disease patient child aspirin resistance data; the method is used for predicting the aspirin resistance of the children with Kawasaki disease, improving the curative effect of the platelet-resisting medicines for the children with Kawasaki disease, and reducing the occurrence of thrombotic complications. The prediction method is quick and convenient, the cost is not required, and the model can be adjusted at any time to adapt to the risk of predicting aspirin resistance in different diseases.
Description
Technical Field
The invention relates to a prediction model and a prediction evaluation system for resistance of Kawasaki disease aspirin, and relates to the technical field of drug resistance prediction.
Background
Kawasaki Disease (KD) is an acute fever eruption disease with systemic vasculitis as the main pathological change, and serious complications of Kawasaki disease comprise coronary aneurysm, coronary thrombosis and even sudden death, and are the main causes of acquired heart disease of children in recent years. Research shows that the activation and the increase of the number of the platelets are high-risk factors of coronary thrombosis, and aspirin is the first choice medicine for platelet resistance treatment in both acute and recovery phases of Kawasaki disease. However, aspirin has great individual difference in antiplatelet effect on children with Kawasaki disease, and aspirin resistance phenomenon exists in the condition that some children take aspirin regularly but the platelet aggregation function cannot be well inhibited. Clinically, whether aspirin resistance exists or not is evaluated mainly by means of thromboelastogram and gene detection, the cost is high, and a simple evaluation means for judging whether aspirin resistance exists or not by inputting clinical data of a child patient is lacked at home and abroad.
Aspirin, as a conventional therapeutic agent, inhibits activation and aggregation of platelets mainly by inhibiting the production of thromboxane A2, and is widely applied to cardiovascular and cerebrovascular diseases and used for preventing thrombotic events. However, patients taking aspirin in large quantities fail to achieve the desired effect and instead develop severe vascular adverse events, a phenomenon known as "aspirin resistance". A great number of aspirin resistance reports in adult cardiovascular disease research, and research shows that aspirin resistance is a meaningful independent index for predicting short-term functional prognosis of patients with ischemic stroke. A meta analysis of the synergistic group of antithrombotic treatments showed that oral antiplatelet drugs for secondary prevention reduced the risk of subsequent myocardial infarction by 25% and mortality by 20% in patients at high risk for cardiovascular events, however, even with secondary prevention, re-hospitalization occurred in about 15% of patients with ischemic heart disease, and a possible reason for such high rate of re-hospitalization was the presence of resistance to aspirin. The presence of thrombotic episodes and aspirin resistance is associated with a significant increase in hospitalized mortality. Studies of Murray et al found that infants with Kawasaki disease do have aspirin resistance, and the dosage of oral aspirin has no correlation with aspirin resistance. At present, no report about aspirin resistance models of children suffering from Kawasaki disease exists at home and abroad.
Current recognition of aspirin resistance relies on the detection of platelet aggregation function. During the 2000-2003 years, many data indicate that some patients have the aspirin resistance determined by platelet function tests, the incidence of aspirin resistance of children suffering from Kawasaki disease is high, and no effective monitoring means exists in the past. In recent years, some scholars evaluate whether aspirin resistance occurs or not through thrombelastogram and gene detection, the cost of the thrombelastogram and the gene detection technology is high, and some parents of children cannot bear the two detections due to economic factors and cannot effectively detect the aspirin resistance phenomenon, so that the occurrence of a thrombus event occurs, the hospitalization frequency and the hospitalization cost are increased, and the economic burden is increased to a certain extent. The gene detection waiting for the detection result probably takes about 20 days to 1 month, and the thromboelastogram can obtain the result in about 4 hours at the fastest speed due to the fact that blood samples need to be detected on a computer, preprocessed and the like. So far, no recognized detection method exists at home and abroad, and whether aspirin resistance occurs in Kawasaki disease children can be immediately predicted by quickly inputting clinical data.
Disclosure of Invention
The invention aims to provide a Kawasaki disease aspirin resistance prediction model and a prediction evaluation system of the Kawasaki disease aspirin resistance prediction model, which are used for predicting the aspirin resistance of children with Kawasaki disease, improving the curative effect of antiplatelet drugs on the children with Kawasaki disease and reducing the occurrence of thrombotic complications. The prediction method is quick and convenient, the cost is not required, and the model can be adjusted at any time to adapt to the risk of predicting aspirin resistance in different diseases.
In order to achieve the technical purpose and achieve the technical effect, the invention is realized by the following technical scheme:
a construction method of a Kawasaki disease aspirin resistance prediction model comprises the following steps:
s1: inputting clinical data of the Kawasaki patient child into Epidata software, and exporting Excel and SPSS files from an Epidata database;
s2: performing data preprocessing including data clearing, data integration and data transformation on the exported SPSS file;
s3: utilizing multi-factor regression analysis to mine high-risk factors resistant to aspirin of children suffering from Kawasaki disease;
the high-risk factors are subjected to regression statistical analysis to discover high-risk factors related to aspirin resistance from indexes such as gender, age, BMI, fever time, IVIG time, withdrawal time after IVIG application, coronary artery dilatation condition, hemoglobin, CRP, blood sedimentation, platelets, albumin, liver function, cytokines, serum lipoprotein, thromboelastogram and the like;
s4: establishing a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model which are digital prediction models of aspirin resistance of four Kawasaki disease children;
s5: analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
s6: and obtaining the analysis result of the digital prediction model, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis result, and obtaining the Kawasaki disease aspirin resistance prediction model.
Further, the variable weight processing of the prediction results of the BP neural network model, the bayesian network model, the decision tree model and the Logistic regression model is as follows: comparing the quality of various models by using a confusion matrix and an ROC curve, evaluating the performance of the models, sequentially giving weights to the models and inputting results;
another object of the present invention is to disclose a predictive evaluation system of a prediction model of resistance to aspirin in kawasaki disease, comprising:
a data acquisition module: the system is used for inputting, importing and collecting clinical data to be tested;
a data analysis module: the data acquisition module is used for acquiring information of a user;
a database module: the system is used for recording historical data for calling, comparing and exporting;
an output module: and the data analysis module is used for outputting the analysis result of the data analysis module.
Further, the data analysis module establishes four digital prediction models of the aspirin resistance of the Kawasaki disease patient child, namely a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model according to the input clinical data;
analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
obtaining the analysis result of the digital prediction model, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis result, and obtaining a Kawasaki disease aspirin resistance prediction model;
and further obtaining prediction data of resistance of the Kawasaki disease aspirin.
Further, the data analysis module obtains the prediction data of the resistance to the aspirin of the Kawasaki disease and also comprises comparison of historical data in the database module.
Furthermore, the prediction evaluation system further comprises a communication module, an interaction module and a power supply module.
Further, the prediction evaluation system is presented in the form of APP and public numbers.
The invention also aims to disclose an application of the prediction evaluation system of the Kawasaki disease aspirin resistance prediction model in prediction of the Kawasaki disease patient aspirin resistance.
The invention has the beneficial effects that:
the invention discloses a Kawasaki disease aspirin resistance prediction model, which is characterized in that a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model are established by data input, the accuracy, the sensitivity, the specificity, the AUC, the positive prediction value and the negative prediction value of four digital prediction models are analyzed by adopting a confusion matrix and an ROC curve, the prediction results of the four digital prediction models are subjected to variable weight processing based on the analysis results, and the Kawasaki disease aspirin resistance prediction model is obtained;
meanwhile, the invention provides a prediction evaluation system of the Kawasaki disease aspirin resistance prediction model, data parameters are collected through a data collection module, the data analysis module establishes the Kawasaki disease aspirin resistance prediction model based on BP neural network model, Bayesian network model, decision tree model and Logistic regression model weight distribution, and the Kawasaki disease aspirin resistance data of children patients are calculated;
the method is used for predicting the aspirin resistance of the children with Kawasaki disease, improving the curative effect of the platelet-resisting medicines for the children with Kawasaki disease, and reducing the occurrence of thrombotic complications. The prediction method is quick and simple, the cost is not required, and the model is adjusted at any time to adapt to the risk of predicting aspirin resistance in different diseases.
Of course, it is not necessary for any product in which the invention is practiced to achieve all of the above-described advantages at the same time.
Drawings
FIG. 1 is a flow chart of a method for constructing a prediction model of Kawasaki disease aspirin resistance according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a BP neural network model according to an embodiment of the present invention;
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail with reference to the embodiments.
Example 1
A construction method of a Kawasaki disease aspirin resistance prediction model comprises the following steps:
s1: inputting clinical data of the Kawasaki disease patient into Epidata software, and exporting Excel and SPSS files from an Epidata database;
s2: performing data preprocessing including data clearing, data integration and data transformation from the exported SPSS file;
s3: utilizing multi-factor regression analysis to mine high-risk factors resistant to aspirin of children suffering from Kawasaki disease; the high-risk factors are subjected to regression statistical analysis to discover high-risk factors related to aspirin resistance from indexes such as gender, age, BMI, fever time, IVIG time, withdrawal time after IVIG, coronary artery dilation condition, hemoglobin, CRP, blood sedimentation, platelets, albumin, liver function, cytokines, serum lipoprotein, thromboelastogram and the like;
s4: establishing a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model as digital prediction models of aspirin resistance of four Kawasaki disease children;
s5: analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
s6: and obtaining the analysis result of the digital prediction model, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis result, and obtaining the Kawasaki disease aspirin resistance prediction model.
Example 2
Based on embodiment 1, as shown in fig. 2, the BP neural network model establishment of the present invention includes: (BP neural network model has higher sensitivity and specificity)
y i Representing output layer neurons, in Z j Representing hidden layer neurons, X k Representing input layer neurons, the weight of input layer neurons K to hidden layer neurons j is ω 1 jk The weight from hidden layer neuron j to output layer neuron i is ω 2 ik Assume that the input layer has K neurons, the hidden layer has J neurons, and the output layer has I neurons. F1 and F2 represent the activation functions of the hidden layer and the output layer, respectively, then the input of the hidden layer j is:
the output is:
the inputs to output unit i are:
the final output is:
for a certain input sample n, the error is defined as:
the overall error function is:
wherein y is i n Represents the predicted output value of the nth sample
Taking clinical indexes related to resistance of aspirin to Kawasaki disease as input parameters of a neural network, wherein the number of output nodes is 1, if the output value is less than or equal to 0.5, the aspirin resistance is determined as non-aspirin resistance, and the output value is greater than 0.5, the aspirin resistance is determined as aspirin resistance; training the BP neural network by using the training set data, and searching the nonlinear relation between the input parameters (clinical indexes) and the output result (whether the input parameters are aspirin resistance or not), so that the BP neural network learns the rule from the input parameters to the output result of the training set, and when the test set data is given, the model diagnoses whether the patient is aspirin resistance or not;
the Bayesian network model:
bayes is based on a total probability formula and is used for describing the relationship between two conditional probabilities, and the probability of complex events is obtained based on the additivity of the conditional probabilities. The formula is as follows:
ai is n pairwise mutually exclusive events (aspirin resistance occurs and aspirin resistance does not occur); p (Ai) is the probability of Ai occurrence in n events, i.e. the prior probability (the proportion of each clinical factor in the total infant patients), and under the condition that each specific clinical factor B of each clinical factor has appeared, the probability of whether each patient has aspirin resistance is calculated according to a formula; B1-Bn are each clinical factor, P (B/Ai) is the probability of occurrence of the clinical factor B when the event Ai occurs, namely the conditional probability; when P (Ai/B) is the occurrence of each clinical factor combination B, the probability of the occurrence of the condition Ai, namely the posterior probability, is compared with the size of P (A1/B) and P (A2/B), if P (A1/B) is obviously larger than P (A2/B), the probability that the patient does not have aspirin resistance is considered to be high, and if P (A2/B) is obviously larger than P (A1/B), the risk that the Sikawasaki patient child has aspirin resistance is suggested to be high. According to the established Bayesian model prior probability and conditional probability, the posterior probability can be calculated, and whether the patient has aspirin resistance or not can be predicted. The Bayesian network is used for representing the probability relation between the clinical information of Kawasaki patients and the occurrence of aspirin resistance, and aspirin resistance is predicted.
A decision tree model:
the decision tree algorithm is a practical CART algorithm, a practical Gini index is taken as an attribute selection standard, and a calculation formula of the Gini index is as follows:
gini (D) represents the uncertainty of the set D, and Gini (D, A) represents the uncertainty of the set D after being divided by the variable A. The larger the kini index, the greater the uncertainty of the sample. According to the research, clinically significant indexes are selected as independent variables and used as internal nodes of a decision tree, whether the asippilin resistance is used as a dependent variable and used as leaf nodes of the decision tree, a decision tree model is trained by using training set data, and a test set is used for inflammation of the model. When testing clinical information of a patient, a decision is made from the root node until a leaf node is encountered that is a prediction of the patient's disease.
Logistic regression model: establishing a Logistic regression model by adopting a stepwise forward regression method and using training set data to a plurality of items with statistical significance to differences in single-factor analysis results, and evaluating the performance of the model through test set data;
four digital prediction models of kawasaki disease children aspirin resistance.
The variable weight processing of the prediction results of the BP neural network model, the Bayesian network model, the decision tree model and the Logistic regression model comprises the following steps: comparing the quality of various models by using a confusion matrix and an ROC curve, evaluating the performance of the models, giving weight to the models and outputting results;
example 3
A predictive evaluation system for a kawasaki disease aspirin resistance predictive model, the predictive evaluation system comprising:
a data acquisition module: the system is used for inputting, importing and collecting clinical data to be tested;
a data analysis module: the data analysis module is used for analyzing and processing the information acquired by the data acquisition module;
a database module: the system is used for recording historical data for calling, comparing and exporting;
an output module: and the data analysis module is used for outputting the analysis result of the data analysis module.
The data analysis module is used for establishing four digital prediction models of the aspirin resistance of the Kawasaki disease patient child, namely a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model according to the input clinical data;
analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
obtaining the analysis result of the digital prediction model, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis result, and obtaining a Kawasaki disease aspirin resistance prediction model;
and further obtaining prediction data of resistance of the Kawasaki disease aspirin.
The data analysis module obtains the prediction data of the resistance of the Kawasaki disease aspirin and also comprises comparison of historical data in the database module.
The prediction evaluation system further comprises a communication module, an interaction module and a power supply module.
The prediction evaluation system appears in the form of APP and public numbers.
Example 4
The application of a prediction evaluation system of a Kawasaki disease aspirin resistance prediction model in prediction of the Kawasaki disease aspirin resistance of children;
aiming at the prediction model, the infant suffering from Kawasaki disease oral aspirin is subjected to detection of thromboelastogram, according to thromboelastogram, aspirin resistance is defined according to the AA inhibition rate of less than or equal to 50%, aspirin sensitivity is positioned according to the AA inhibition rate of more than 50%, children suffering from Kawasaki disease are divided into an aspirin resistance group and an aspirin sensitivity group, meanwhile, 3 years of clinical data (including sex, age, BMI, fever time, IVIG retroversion time, coronary artery dilatation condition, hemoglobin, CRP, blood sedimentation, platelets, albumin, liver function, cytokines, serum lipoprotein and thromboelastogram indexes) are collected, the clinical data are input into Epidata software, Excel and SPSS files are exported, and data preprocessing is performed, including data clearing, data integration and data transformation; through the detection of the thromboelastogram of the Kawasaki disease children at a single center, the aspirin resistance rate of the Kawasaki disease children taking aspirin is about 30%, the incidence rate of the aspirin resistance in the aspirin resistance children is 3%, the incidence rate of the aspirin resistance of the children in a more sensitive group is high, and the optimal treatment time is delayed due to the delay of the thromboelastogram detection to cause the occurrence of the thrombosis event;
digging high-risk factors resistant to aspirin of the Kawasaki disease patient by utilizing multi-factor regression analysis, and establishing four digital prediction models of the Kawasaki disease patient aspirin resistance, namely a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model;
analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
and obtaining the analysis result of the digital prediction model, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis result, and obtaining the Kawasaki disease aspirin resistance prediction model.
The method is used for predicting the aspirin resistance of the children with Kawasaki disease, improving the curative effect of the platelet-resisting medicines for the children with Kawasaki disease, and reducing the occurrence of thrombotic complications. The prediction method is quick and simple, the cost is not required, and the model can be adjusted at any time to adapt to different diseases to predict the risk of aspirin resistance.
In the description herein, references to the description of "one embodiment," "an example," "a specific example," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand the invention for and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (9)
1. A construction method of a Kawasaki disease aspirin resistance prediction model is characterized by comprising the following steps of:
s1: inputting clinical data of the Kawasaki patient child into Epidata software, and exporting Excel and SPSS files from an Epidata database;
s2: performing data preprocessing including data clearing, data integration and data transformation from the exported SPSS file;
s3: digging high-risk factors of aspirin resistance of children suffering from Kawasaki disease by utilizing multi-factor regression analysis, wherein the high-risk factors are obtained by digging indexes related to aspirin resistance through regression statistical analysis;
s4: establishing four digital prediction models of the aspirin resistance of the Kawasaki disease patient, namely a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model;
s5: analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
s6: and obtaining the analysis result of the digital prediction model, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis result, and obtaining the Kawasaki disease aspirin resistance prediction model.
2. The method of constructing a prediction model of kawasaki disease aspirin resistance as claimed in claim 1, wherein: the high-risk factors of the kawasaki disease infant aspirin resistance are high-risk factors related to aspirin resistance discovered by carrying out regression statistical analysis on indexes such as gender, age, BMI, fever time, IVIG application time, IVIG post-withdrawal time, coronary artery dilatation conditions, hemoglobin, CRP, blood sedimentation, platelets, albumin, liver functions, cytokines, serum lipoproteins, thromboelastogram and the like.
3. The method of constructing a prediction model of kawasaki disease aspirin resistance as claimed in claim 2, wherein: the variable weight processing of the prediction results of the BP neural network model, the Bayesian network model, the decision tree model and the Logistic regression model comprises the following steps: and comparing the advantages and the disadvantages of various models by using the confusion matrix and the ROC curve, evaluating the performance of the models, sequentially giving weights to the models and inputting the results.
4. A prediction evaluation system of a Kawasaki disease aspirin resistance prediction model is characterized in that:
the predictive evaluation system includes:
a data acquisition module: the system is used for inputting, importing and collecting clinical data to be tested;
a data analysis module: the data analysis module is used for analyzing and processing the information acquired by the data acquisition module;
a database module: the system is used for recording historical data for calling, comparing and exporting;
an output module: and the data analysis module is used for outputting the analysis result of the data analysis module.
5. The system of claim 4, wherein the system comprises: the data analysis module is used for establishing four digital prediction models of the aspirin resistance of the Kawasaki disease patient child, namely a BP neural network model, a Bayesian network model, a decision tree model and a Logistic regression model according to the input clinical data;
analyzing the Accuracy (Accuracy), Sensitivity (Sensitivity), Specificity (Specificity), AUC (area Under Roc), Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of the four digital prediction models by using a confusion matrix and an ROC curve;
obtaining the analysis results of the digital prediction models, performing variable weight processing on the prediction results of the four digital prediction models based on the analysis results, and obtaining a Kawasaki disease aspirin resistance prediction model;
and further obtaining prediction data of resistance of the Kawasaki disease aspirin.
6. The system of claim 5, wherein the system comprises: the data analysis module obtains the prediction data of the Kawasaki disease aspirin resistance and also comprises comparison of historical data in the database module.
7. The system of claim 4, wherein the system comprises: the prediction evaluation system further comprises a communication module, an interaction module and a power supply module.
8. The system of claim 4, wherein the system comprises: the prediction evaluation system appears in the form of APP and public numbers.
9. Use of the prediction evaluation system of the Kawasaki disease aspirin resistance prediction model of claim 4 in prediction of Kawasaki disease patients' aspirin resistance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210359448.7A CN115101217A (en) | 2022-04-06 | 2022-04-06 | Kawasaki disease aspirin resistance prediction model and prediction evaluation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210359448.7A CN115101217A (en) | 2022-04-06 | 2022-04-06 | Kawasaki disease aspirin resistance prediction model and prediction evaluation system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115101217A true CN115101217A (en) | 2022-09-23 |
Family
ID=83287870
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210359448.7A Withdrawn CN115101217A (en) | 2022-04-06 | 2022-04-06 | Kawasaki disease aspirin resistance prediction model and prediction evaluation system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115101217A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115689298A (en) * | 2022-12-30 | 2023-02-03 | 北京码牛科技股份有限公司 | Telecommunication fraud risk prediction method, system, equipment and readable storage medium |
-
2022
- 2022-04-06 CN CN202210359448.7A patent/CN115101217A/en not_active Withdrawn
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115689298A (en) * | 2022-12-30 | 2023-02-03 | 北京码牛科技股份有限公司 | Telecommunication fraud risk prediction method, system, equipment and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Du et al. | Accurate prediction of coronary heart disease for patients with hypertension from electronic health records with big data and machine-learning methods: model development and performance evaluation | |
Shaji | Predictionand diagnosis of heart disease patients using data mining technique | |
Joo et al. | Clinical implication of machine learning in predicting the occurrence of cardiovascular disease using big data (Nationwide Cohort Data in Korea) | |
Yao et al. | Liver disease screening based on densely connected deep neural networks | |
Nikam et al. | Cardiovascular disease prediction using machine learning models | |
Khan et al. | A Comparative Study of Machine Learning classifiers to analyze the Precision of Myocardial Infarction prediction | |
Khatatneh et al. | Hybrid neural networks with virtual flows in in medical risk classifiers | |
Shamsollahi et al. | Using combined descriptive and predictive methods of data mining for coronary artery disease prediction: a case study approach | |
Nasajpour et al. | Federated transfer learning for diabetic retinopathy detection using CNN architectures | |
Annamalai et al. | Diabetes mellitus prediction and severity level estimation using OWDANN algorithm | |
CN115101217A (en) | Kawasaki disease aspirin resistance prediction model and prediction evaluation system | |
Adi et al. | Stroke risk prediction model using machine learning | |
Li et al. | [Retracted] Prediction Model of Ischemic Stroke Recurrence Using PSO‐LSTM in Mobile Medical Monitoring System | |
Hassani et al. | An approach of predicting heart disease using a hybrid neural network and decision tree | |
De Baets et al. | Positive blood culture detection in time series data using a BiLSTM network | |
Tang et al. | Disease network delineates the disease progression profile of cardiovascular diseases | |
Deepa et al. | Experimental evaluation of artificial intelligence assisted heart disease prediction using deep learning principle | |
Tahsin et al. | Predictive analysis & brief study of early-stage diabetes using multiple classifier models | |
Singh et al. | Classification of arrhythmia using transfer learning and its applications in smart wearables | |
Schwarzerova et al. | An innovative perspective on metabolomics data analysis in biomedical research using concept drift detection | |
Jadhav et al. | Monitoring and Predicting of Heart Diseases Using Machine Learning Techniques | |
Chanthaweethip et al. | Temporal data mining and visualization for treatment outcome prediction in HIV patients | |
Shankhdhar | Visualization and Prediction of Heart Disease using Big Data Analytics | |
Mamman et al. | Diabetic classification using cascaded data mining technique | |
Kumar et al. | Cardiac diagnostic feature and demographic identification models: a futuristic approach for smart healthcare using machine learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220923 |
|
WW01 | Invention patent application withdrawn after publication |