Auxiliary diagnosis method and system for predicting coronary heart disease based on machine learning, medium and processor

By using machine learning-based methods and training a multi-layer model to develop a coronary heart disease risk scoring model, the problem of insufficient data information in existing technologies is solved, thereby improving the accuracy and efficiency of coronary heart disease diagnosis and providing customized risk assessment.

CN119049688APending Publication Date: 2024-11-29THE SECOND AFFILIATED HOSPITAL OF GUANGXI MEDICAL UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411143412.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

In existing technologies, the training data for coronary heart disease prediction models consists only of peripheral blood core cells, which has limited information and results in low prediction accuracy.

Method used

A machine learning-based approach was adopted to extract multiple key features from historical data of coronary heart disease patients and non-coronary heart disease patients recorded in hospitals, forming a sample set. A two-layer machine learning model was then used for training to form a coronary heart disease risk scoring model, which includes a combination of random forest model, gradient boosting tree model and neutral neural network model.

Benefits of technology

It improves the accuracy and efficiency of coronary heart disease diagnosis, reduces the subjectivity and uncertainty of doctors in the diagnostic process, provides customized risk assessment, and helps doctors understand patients' coronary heart disease risk more accurately.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049688A_ABST
    Figure CN119049688A_ABST
Patent Text Reader

Abstract

The invention discloses an auxiliary diagnosis method and system for predicting coronary heart disease based on machine learning, a medium and a processor, and relates to the technical field of statistical prediction, and the method comprises the steps: obtaining the historical data of patients recorded in a hospital, including the historical data of all coronary heart disease patients and the historical data of non-coronary heart disease patients seeing a doctor in the department of cardiology; extracting multiple key features related to the symptoms of the coronary heart disease in the historical data, and forming a sample set by taking the multiple key features of each patient as a sample; dividing the sample set into a training set and a verification set according to a preset proportion, and inputting the training set and the verification set into a two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score; and inputting multiple key features of the patient to be treated into the coronary heart disease prediction model to obtain a coronary heart disease risk score of the patient to be treated. According to a large amount of historical data of the patient, the subjectivity of the doctor in the diagnosis process is reduced based on machine learning auxiliary diagnosis, and the diagnosis efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of statistical prediction, and in particular to an auxiliary diagnosis method, system, medium and processor for predicting coronary heart disease based on machine learning. Background Art

[0002] Coronary artery disease (CAD) is one of the diseases that causes the most deaths worldwide. As a type of cardiovascular disease, the number of deaths caused by CAD is also on the rise.

[0003] Many tests may help diagnose coronary heart disease, including: electrocardiogram, cardiac stress test, coronary artery computed tomography and coronary angiography, among which coronary angiography is the main basis for diagnosis. However, in general, coronary angiography will only be carried out when symptoms appear and are very serious. Therefore, there is currently a lack of simple and effective means of screening for coronary heart disease. In the prior art, the training method, computer equipment and readable storage medium of the coronary heart disease prediction model with patent number CN117637174A establish prediction models for different stages of coronary heart disease progression and reduce the incidence of adverse cardiovascular events. However, the data used for model training in this scheme is only peripheral blood core cells, the amount of information contained is relatively limited, and the prediction accuracy of coronary heart disease is not high. Summary of the invention

[0004] In view of the problem that the model training data for establishing a prediction model for coronary heart disease in the prior art is only peripheral blood core cells, the amount of information contained is relatively limited, and the prediction accuracy of coronary heart disease is not high, the present invention provides an auxiliary diagnosis method, system, medium and processor for predicting coronary heart disease based on machine learning, which can reduce the subjectivity and uncertainty of doctors in the diagnosis process and improve the diagnosis efficiency according to machine learning auxiliary diagnosis. The specific technical scheme is as follows:

[0005] An auxiliary diagnosis method for predicting coronary heart disease based on machine learning comprises the following steps:

[0006] Obtain the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department;

[0007] Extracting multiple key features related to coronary heart disease symptoms from the historical data, and taking the multiple key features of each patient as a sample to form a sample set;

[0008] Dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score;

[0009] A plurality of key characteristics of the patient are input into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient.

[0010] Preferably, dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score comprises:

[0011] The training set is subjected to hyperparameter optimization using a random forest model and a gradient boosted tree model, respectively, and the optimal hyperparameter combination is obtained based on cross-validation of the training set;

[0012] The random forest model and the gradient boosting tree model were trained using the optimal hyperparameter combination, and the random forest model and the gradient boosting tree model were configured to output the probability of coronary heart disease for each sample;

[0013] The probability of coronary heart disease of each sample is input into the neutral neural network machine as the posterior probability for hyperparameter optimization to obtain a trained neutral neural network model;

[0014] The random forest model and the gradient boosting tree model are used as the first layer of the coronary heart disease prediction model, and the neutral neural network model is used as the second layer of the coronary heart disease prediction model to obtain a coronary heart disease prediction model based on two-layer machine learning.

[0015] Preferably, the step of inputting a plurality of key features of the patient into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient includes:

[0016] A plurality of key features of the patient are input into the coronary heart disease prediction model, and the posterior probability calculated by the neutral neural network model of the coronary heart disease prediction model is used as the coronary heart disease risk score.

[0017] Preferably, the obtaining of the historical data of the patient recorded in the hospital specifically refers to:

[0018] Obtain the historical data of all patients with coronary heart disease recorded in the hospital and the historical data of patients without coronary heart disease who visited the cardiology department. The historical data include laboratory test data, medical device test data, medical history data, family medical history data and basic patient information data.

[0019] Preferably, extracting a plurality of key features related to coronary heart disease symptoms from the historical data and taking the plurality of key features of each patient as a sample to form a sample set comprises:

[0020] dividing the historical data into data in the form of categorical data, text data and numerical data;

[0021] Through data conversion, both text data and digital data are converted into classified data;

[0022] The converted classified data are filtered to obtain multiple key features related to the symptoms of coronary heart disease, and the multiple key features of each patient are taken as a sample to form a sample set.

[0023] Preferably, an auxiliary diagnosis method for predicting coronary heart disease based on machine learning also includes:

[0024] The coronary heart disease risk score of the patient is compared with a preset risk threshold, and the examination items of the patient are output according to the comparison result.

[0025] An auxiliary diagnosis system for predicting coronary heart disease based on machine learning, applied to the aforementioned auxiliary diagnosis method for predicting coronary heart disease based on machine learning, comprising:

[0026] A data acquisition unit, used to acquire the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department;

[0027] A feature extraction unit, used to extract multiple key features related to coronary heart disease symptoms from the historical data, and take the multiple key features of each patient as a sample to form a sample set;

[0028] A model learning generation unit, used for dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score;

[0029] The disease prediction unit is used to input multiple key features of the patient into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient.

[0030] A computer-readable storage medium, comprising a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the aforementioned auxiliary diagnosis method for predicting coronary heart disease based on machine learning.

[0031] A processor is used to run a program, wherein when the program is run, the aforementioned auxiliary diagnosis method for predicting coronary heart disease based on machine learning is executed.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The auxiliary diagnosis method for predicting coronary heart disease based on machine learning of the present invention obtains the historical data of patients recorded in the hospital, including the historical data of all coronary heart disease patients and the historical data of non-coronary heart disease patients who visited the cardiology department; extracts multiple key features related to the symptoms of coronary heart disease in the historical data, and uses the multiple key features of each patient as a sample to form a sample set; divides the sample set into a training set and a validation set according to a preset ratio, and inputs a two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score; inputs the multiple key features of the visiting patients into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the visiting patients. The present invention automatically processes a large amount of data, and the auxiliary diagnosis based on machine learning can reduce the subjectivity and uncertainty of doctors in the diagnosis process and improve the efficiency of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0035] Figure 1 This is a flow chart of an auxiliary diagnosis method for predicting coronary heart disease based on machine learning according to the present invention.

[0036] Figure 2 This is a schematic diagram of an auxiliary diagnosis system for predicting coronary heart disease based on machine learning according to the present invention.

[0037] Figure 3 This is a schematic diagram of the coronary heart disease prediction results of an embodiment of the two-layer machine learning coronary heart disease prediction model of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0040] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0041] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0042] See also Figure 1 The present application provides an auxiliary diagnosis method for predicting coronary heart disease based on machine learning, comprising the following steps:

[0043] Step S01, obtaining the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department;

[0044] The historical data of all patients with coronary heart disease recorded in the hospital and the historical data of patients without coronary heart disease who visited the cardiology department were obtained. The historical data included laboratory test data (such as blood routine, kidney routine, liver routine, etc.), medical device test data (such as electrocardiogram, ultrasound, etc.), medical history data (such as drinking history, smoking history, etc.), family medical history data (family medical history, etc.), basic information of patients (height, gender, weight, place of origin, etc.), etc., totaling 167 items of data.

[0045] Step S02, extracting multiple key features related to coronary heart disease symptoms from the historical data, and taking the multiple key features of each patient as a sample to form a sample set;

[0046] The historical data is divided into data in the form of categorical data (such as gender), text data (such as medical history description) and digital data (such as blood pressure); the text data and digital data are converted into categorical data through data conversion; the converted categorical data is filtered to obtain multiple key features related to coronary heart disease symptoms (96 items in this embodiment), and the multiple key features of each patient are taken as a sample to form a sample set.

[0047] Step S03, dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score;

[0048] In this embodiment, the sample set is divided into a training set and a validation set at a ratio of 80% and 20%, and the training set is subjected to hyperparameter optimization using a random forest model and a gradient boosted tree model, respectively, and the optimal hyperparameter combination is obtained based on cross-validation of the training set;

[0049] The basic principle of cross validation is to group the sample set, use a part of the data in each group as the training set to train the model, and use the rest as the validation set to test the model, which is used as the performance evaluation indicator of the model. Cross validation means dividing the sample set into n groups of approximately equal size, and then using each subset of the n groups as a validation set, and the remaining n-1 subsets as training sets to train the model. The cross validation is repeated n times, and the n test results are averaged to finally get an estimated value.

[0050] The random forest model and the gradient boosting tree model were trained using the optimal hyperparameter combination, and the random forest model and the gradient boosting tree model were configured to output the probability of coronary heart disease for each sample;

[0051] The probability of coronary heart disease of each sample is input into the neutral neural network machine as the posterior probability for hyperparameter optimization to obtain a trained neutral neural network model;

[0052] The random forest model and the gradient boosting tree model are used as the first layer of the coronary heart disease prediction model, and the neutral neural network model is used as the second layer of the coronary heart disease prediction model to obtain a coronary heart disease prediction model based on two-layer machine learning.

[0053] Use the test set to verify whether the accuracy of the coronary heart disease prediction model meets the requirements. If it does not meet the requirements, I need to re-tune the parameters of the random forest model and gradient boosting tree model in the first layer of the model, and perform learning and training, as well as re-tune the parameters of the neutral neural network model in the second layer of the model, and perform learning and training until the model meets the test accuracy requirements.

[0054] Step S04: input multiple key features of the patient into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient.

[0055] In this embodiment, multiple key features of the patient are input into the coronary heart disease prediction model, and the posterior probability calculated by the neutral neural network model of the coronary heart disease prediction model is used as the coronary heart disease risk score.

[0056] It should be noted that the historical data recorded by the hospital is the data generated by the patient's examination, including outpatient visits and examination data. Suspected patients in the cardiology outpatient clinic complete the patient's physical examination through the examination form issued by the outpatient clinic, including medical instrument examination, laboratory examination, medical history inquiry, basic situation inquiry, etc., to form a series of health data.

[0057] In this embodiment, both text data and digital data are converted into classified data through a data conversion method. The specific implementation method is: for example, in the original classified data, in "gender", "Male" and "Female" are encoded as "1" and "0" respectively; the text data is converted into classified data using the same method, and in "whether suffering from asthma", "whether there are serious difficulties in walking or climbing stairs", "whether a heavy drinker", "Yes" and "No" are encoded as "1" and "0" classification forms respectively. The same principle applies to digital data. In "whether blood pressure exceeds the threshold", "Yes" and "No" are encoded as "1" and "0" classification forms. By converting digital data and text data into classified data, the computer's operation recognition efficiency is improved.

[0058] After the data is converted, it is filtered and individuals or data items with too much missing data (set to >30% based on practical experience) are deleted. The filtered data (filtered data) still has some missing data, and the data is filled in through the random forest missing data filling algorithm to make the data complete and improve the data quality. The missing data filling algorithms involved here include the average algorithm (using the average of the remaining sample data as the missing value), the similarity interpolation method (calculating the similarity between missing samples and non-missing samples, and then estimating the missing value through the weighted value of the non-missing sample), and the MissForest prediction filling method (finding the intrinsic correlation between the non-missing data column and the missing data column, predicting and filling the missing data through the non-missing data), etc.

[0059] In this embodiment, the 96 key features selected are not only automatically filtered, but also comprehensively selected based on expert clinical experience. The 96 key features select the feature subset that has the greatest impact on model performance, improve model accuracy, and reduce running time. The purpose of feature selection is to train a model with the best comprehensive indicators using as few features as possible.

[0060] In another embodiment, a two-layer machine learning-assisted diagnosis method for predicting coronary heart disease using admission data, wherein the prediction results of the two-layer machine learning coronary heart disease prediction model are as follows: Figure 3 The receiver operating characteristic curve area (AUROC) of the double-layer machine learning coronary heart disease prediction model for predicting coronary heart disease risk can reach 0.88, and the prediction accuracy can reach above 0.8 (sample size 2400 people). This result shows that the use of the double-layer machine learning coronary heart disease prediction model to predict coronary heart disease risk can achieve better prediction results and has better practical effects.

[0061] The auxiliary diagnosis method for predicting coronary heart disease based on machine learning of the present invention obtains the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department; extracts multiple key features related to the symptoms of coronary heart disease in the historical data, and uses the multiple key features of each patient as a sample to form a sample set; divides the sample set into a training set and a validation set according to a preset ratio, and inputs a two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score; inputs the multiple key features of the visiting patients into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the visiting patients. The present invention automatically processes a large amount of data, and the auxiliary diagnosis based on machine learning can reduce the subjectivity and uncertainty of doctors in the diagnosis process and improve the diagnostic efficiency. At the same time, the prediction model based on the two-layer machine learning can provide customized risk assessment for each patient, helping doctors to more accurately understand the patient's coronary heart disease risk level.

[0062] Specifically, in a preferred embodiment of the present application, an auxiliary diagnosis method for predicting coronary heart disease based on machine learning also includes:

[0063] The coronary heart disease risk score of the patient is compared with a preset risk threshold, and the examination items of the patient are output according to the comparison result.

[0064] In specific implementation, when the patient's coronary heart disease risk score is greater than or equal to the preset risk threshold, it means that the patient has a high probability of suffering from coronary heart disease, and coronary angiography is needed to prove whether the coronary heart disease is confirmed; when the patient's coronary heart disease risk score is less than the preset risk threshold, the patient's coronary heart disease risk can be ruled out and other examinations can be arranged.

[0065] See also Figure 2 The embodiment of the present application further provides an auxiliary diagnosis system for predicting coronary heart disease based on machine learning, which is applied to the aforementioned auxiliary diagnosis method for predicting coronary heart disease based on machine learning, including:

[0066] A data acquisition unit, used to acquire the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department;

[0067] A feature extraction unit, used to extract multiple key features related to coronary heart disease symptoms from the historical data, and take the multiple key features of each patient as a sample to form a sample set;

[0068] A model learning generation unit, used for dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score;

[0069] The disease prediction unit is used to input multiple key features of the patient into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient.

[0070] In this embodiment, the functional explanation corresponding to each unit is the same as the explanation of the aforementioned method for constructing a carbon assessment classification index system for a zero-carbon power supply station, and the technical effect is also the same as the technical effect of an auxiliary diagnosis method for predicting coronary heart disease based on machine learning in the embodiment, and will not be repeated here.

[0071] An embodiment of the present application also provides a computer-readable storage medium, which includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to execute the aforementioned auxiliary diagnosis method based on machine learning to predict coronary heart disease.

[0072] The technical effect of this embodiment is the same as the technical effect of an auxiliary diagnosis method for predicting coronary heart disease based on machine learning in an embodiment, and will not be repeated here.

[0073] The present invention can be used in many general or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.

[0074] An embodiment of the present application also provides a processor, which is used to run a program, wherein when the program is run, the aforementioned auxiliary diagnosis method for predicting coronary heart disease based on machine learning is executed.

[0075] The technical effect of this embodiment is the same as the technical effect of the auxiliary diagnosis method for predicting coronary heart disease based on machine learning in embodiment 1, and will not be repeated here.

[0076] The processor in this embodiment may be a central processing unit (CPU), a controller, a microcontroller, or other data processing chips.

[0077] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0078] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0079] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0080] If the integrated unit is implemented in the form of 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 the present invention, 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, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0081] Finally, 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. An auxiliary diagnosis method for predicting coronary heart disease based on machine learning, characterized in that: The following steps are involved: Obtain the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department; Extracting multiple key features related to coronary heart disease symptoms from the historical data, and taking the multiple key features of each patient as a sample to form a sample set; Dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score; A plurality of key characteristics of the patient are input into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient.

2. The auxiliary diagnosis method for predicting coronary heart disease based on machine learning according to claim 1, characterized in that: The method of dividing the sample set into a training set and a validation set according to a preset ratio and inputting the sample set into a two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score includes: The training set is subjected to hyperparameter optimization using a random forest model and a gradient boosted tree model, respectively, and the optimal hyperparameter combination is obtained based on cross-validation of the training set; The random forest model and the gradient boosting tree model were trained using the optimal hyperparameter combination, and the random forest model and the gradient boosting tree model were configured to output the probability of coronary heart disease for each sample; The probability of coronary heart disease of each sample is input into the neutral neural network machine as the posterior probability for hyperparameter optimization to obtain a trained neutral neural network model; The random forest model and the gradient boosting tree model are used as the first layer of the coronary heart disease prediction model, and the neutral neural network model is used as the second layer of the coronary heart disease prediction model to obtain a coronary heart disease prediction model based on two-layer machine learning.

3. The auxiliary diagnosis method for predicting coronary heart disease based on machine learning according to claim 2, characterized in that: The step of inputting multiple key features of the patient into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient includes: inputting multiple key features of the patient into the coronary heart disease prediction model, and using the posterior probability calculated by the neutral neural network model of the coronary heart disease prediction model as the coronary heart disease risk score.

4. The auxiliary diagnosis method for predicting coronary heart disease based on machine learning according to claim 1, characterized in that: The acquisition of historical data of patients recorded in the hospital specifically refers to: Obtain the historical data of all patients with coronary heart disease recorded in the hospital and the historical data of patients without coronary heart disease who visited the cardiology department. The historical data include laboratory test data, medical device test data, medical history data, family medical history data and basic patient information data.

5. The auxiliary diagnosis method for predicting coronary heart disease based on machine learning according to claim 4, characterized in that: The extracting of multiple key features related to coronary heart disease symptoms from the historical data and taking the multiple key features of each patient as a sample to form a sample set includes: dividing the historical data into data in the form of categorical data, text data and numerical data; Through data conversion, both text data and digital data are converted into classified data; The converted classified data are filtered to obtain multiple key features related to the symptoms of coronary heart disease, and the multiple key features of each patient are taken as a sample to form a sample set.

6. The auxiliary diagnosis method for predicting coronary heart disease based on machine learning according to claim 1, characterized in that: Also includes: The coronary heart disease risk score of the patient is compared with a preset risk threshold, and the examination items of the patient are output according to the comparison result.

7. An auxiliary diagnosis system for predicting coronary heart disease based on machine learning, characterized in that: The auxiliary diagnosis method for predicting coronary heart disease based on machine learning as described in any one of claims 1 to 6 comprises: A data acquisition unit, used to acquire the historical data of patients recorded in the hospital, including the historical data of all patients with coronary heart disease and the historical data of patients without coronary heart disease who visited the cardiology department; A feature extraction unit, used to extract multiple key features related to coronary heart disease symptoms from the historical data, and take the multiple key features of each patient as a sample to form a sample set; A model learning generation unit, used for dividing the sample set into a training set and a validation set according to a preset ratio, and inputting the two-layer machine learning model for learning to form a coronary heart disease prediction model with an output coronary heart disease risk score; The disease prediction unit is used to input multiple key features of the patient into the coronary heart disease prediction model to obtain the coronary heart disease risk score of the patient.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the auxiliary diagnosis method for predicting coronary heart disease based on machine learning as described in any one of claims 1 to 6.

9. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the auxiliary diagnosis method for predicting coronary heart disease based on machine learning as described in any one of claims 1 to 6.

Citation Information

Cited By

  • Coronary heart disease prediction method and system based on width learning and survival analysis

    CN121393858A