Myocardial infarction prediction data acquisition method and device and readable storage medium
Through the method of obtaining myocardial infarction prediction data based on simulated annealing algorithm, the prediction model of myocardial infarction was constructed and optimized, and the problem of inaccurate prediction of the existing central infarction detection method was solved, and the timely and accurate prediction of myocardial infarction was achieved. and reliability.
Patent Information
- Application Number
- CN202411000107.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-29
AI Technical Summary
The existing myocardial infarction detection methods are difficult to provide accurate predictions and timely send them to the hospital when there are no obvious signs in the early stage of the onset of myocardial infarction, which makes it difficult for patients to obtain effective rescue and is prone to misjudgment.
Using the method of obtaining myocardial infarction prediction data based on simulated annealing algorithm, a prediction model of myocardial infarction is established by constructing multiple characteristic parameters and myocardial infarction prediction features, and using the copy exchange simulation annealing algorithm to obtain optimized characteristic parameters to build accurate myocardial infarction prediction Model.
It achieves timely and accurate prediction of the occurrence of myocardial infarction, reduces the rate of misjudgment, and improves the accuracy and reliability of myocardial infarction prediction.
Smart Images

Figure CN119049704A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a method, device and readable storage medium for acquiring myocardial infarction prediction data based on a simulated annealing algorithm. Background Art
[0002] Cardiovascular disease is an important risk factor for vascular morbidity and mortality. Among cardiovascular diseases that may endanger life, stroke and myocardial infarction are the most dangerous types of cardiovascular disease. It is particularly important that myocardial infarction (hereinafter referred to as MI) is caused by acute myocardial ischemic necrosis, which leads to rapid onset of MI and difficulty in rapid and accurate prediction.
[0003] Most of the existing methods for acquiring myocardial infarction detection data focus on the in-process diagnosis of myocardial infarction in patients. For example, the prior art provides a smart bracelet, which is connected to a cloud-based evaluation system through a communication module. The smart bracelet obtains the user's daily physiological signs data; then the above physiological signs data are uploaded to the cloud-based evaluation system through the communication module. When a user has a myocardial infarction, the user's heartbeat data is obtained through the user's electrocardiogram and classified based on the heartbeat as a feature. By combining the user's daily physiological signs and the user's heartbeat data in the above manner, it is determined whether a myocardial infarction has occurred or the location data of the myocardial infarction is given.
[0004] However, the above method combines the user's daily physiological signs data and heart rate data to determine whether there is a myocardial infarction or myocardial infarction location data. The amount of daily physiological data is large and diverse, and it is difficult to determine whether it is strongly correlated with myocardial infarction; heart rate data is not specifically for myocardial infarction prediction, which leads to inaccurate prediction of myocardial infarction data by existing methods, and even causes misjudgment.
[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention
[0006] The main purpose of this application is to provide a method, device and readable storage medium for acquiring myocardial infarction prediction data based on a simulated annealing algorithm, aiming to solve the technical problem that the existing methods are inaccurate in predicting myocardial infarction data and may even easily cause misjudgment.
[0007] To achieve the above objectives, the present application proposes a method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm, comprising: Construct multiple characteristic parameters for myocardial infarction prediction; A myocardial infarction prediction model is established based on multiple characteristic parameters combined with corresponding myocardial infarction prediction features and weight factors; From the physical sign data of myocardial infarction patients, a data set of various myocardial infarction prediction features was obtained by screening according to the random sampling method; Select the characteristic data in the data set, input it into the myocardial infarction prediction model, and use the replica exchange simulated annealing algorithm to solve and obtain the optimized characteristic parameters; A myocardial infarction prediction model including optimized feature parameters is used to predict the results of the myocardial infarction prediction features to be tested, thereby obtaining myocardial infarction prediction data.
[0008] Preferably, in the above-mentioned method for acquiring myocardial infarction prediction data, the step of constructing a plurality of characteristic parameters for myocardial infarction prediction includes: Acquire a data feature set of multiple myocardial infarction prediction features, the data feature set including myocardial infarction prediction features and multiple associated feature parameters; For each myocardial infarction prediction feature, the characteristic parameters and the parameter-characteristic curves of the corresponding myocardial infarction prediction features were established by the fixed variable method; Perform curve fitting on the parameter-characteristic curve diagram to obtain the correlation between each characteristic parameter and the corresponding myocardial infarction prediction feature; In descending order of correlation, the top N feature parameters that are strongly correlated with the myocardial infarction prediction features are selected and set as correlation parameters; Using the associated parameters, multiple characteristic parameters are constructed.
[0009] Preferably, in the above-mentioned method for acquiring myocardial infarction prediction data, the step of establishing a myocardial infarction prediction model according to a plurality of characteristic parameters combined with corresponding myocardial infarction prediction features and weight factors comprises: Using multiple feature parameters, combined with corresponding myocardial infarction prediction features and weight factors, the expression of myocardial infarction prediction model is constructed:
[0010] in, Represents the total myocardial infarction energy, represents the feature energy corresponding to the i-th myocardial infarction prediction feature, Indicates j Item weight factor, Indicates i Item characteristic parameters, Indicates j Myocardial infarction prediction features; According to the energy difference between the model calculated value and the actual value of the myocardial infarction energy, the loss function of the myocardial infarction prediction model is constructed.
[0011] Preferably, in the above-mentioned method for acquiring myocardial infarction prediction data, the step of selecting myocardial infarction prediction feature data in the data set and inputting it into the myocardial infarction prediction model comprises: Converting a data set of multiple myocardial infarction prediction features into a multi-dimensional prediction feature map; Randomly select any position coordinate point from the multi-dimensional prediction feature map; Using a K-nearest neighbor algorithm, a set of myocardial infarction prediction feature sets within a predetermined distance range is selected with the position coordinate point as the center, wherein the set of myocardial infarction prediction feature sets includes the above-mentioned multiple myocardial infarction prediction features; The above-mentioned multiple myocardial infarction prediction features are input into the myocardial infarction prediction model for solution; Re-execute the random selection of any position point from the multi-dimensional prediction feature map and subsequent steps until the multi-dimensional prediction feature map is traversed.
[0012] Preferably, in the above-mentioned method for obtaining myocardial infarction prediction data, the step of using a replica exchange simulated annealing algorithm to solve and obtain optimized characteristic parameters includes: A parameter optimization framework corresponding to the replica exchange simulated annealing algorithm is configured, wherein the parameter optimization framework includes a recursive loop, an exchange loop, and a scanning loop; Generate multiple parameter copies, where each parameter copy represents a data set containing multiple sets of feature parameters, and each set of feature parameters is a set of multiple feature parameters; In the recursive loop, the simulated annealing algorithm is executed in parallel using multiple copies of the parameters; In the exchange cycle, the characteristic parameters in two adjacent parameter copies are exchanged according to a predetermined temperature exchange strategy; In the scanning cycle, multiple characteristic parameters and characteristic data are used to execute the simulated annealing algorithm and the energy difference of the myocardial infarction prediction model is calculated using the loss function; A set of characteristic parameters corresponding to the minimum energy difference is selected as the optimized characteristic parameters.
[0013] Preferably, in the above-mentioned method for acquiring myocardial infarction prediction data, the step of using a myocardial infarction prediction model including optimized feature parameters to predict the result of the myocardial infarction prediction feature to be tested comprises: Real-time monitoring of multiple myocardial infarction prediction features of the user to be tested; Inputting a variety of myocardial infarction prediction features into the optimized myocardial infarction prediction model to obtain myocardial infarction prediction data; According to the risk level of the myocardial infarction prediction data, a corresponding warning signal and presentation method are selected, and the myocardial infarction prediction data is sent.
[0014] Preferably, the method for obtaining myocardial infarction prediction data further comprises, after the step of using the myocardial infarction prediction model including the optimized feature parameters to predict the result of the myocardial infarction prediction feature to be tested to obtain the myocardial infarction prediction data: Counting the model prediction value of the infarction prediction model at the predetermined time; The Euclidean distance between the model-predicted value and the actual value of MI was calculated; The Euclidean distance was used to calculate the similarity between the model prediction value and the actual value of myocardial infarction, and the myocardial infarction prediction model was scored according to the similarity; When the score of the myocardial infarction prediction model is less than or equal to the predetermined score threshold, the replica exchange simulated annealing algorithm is reused to solve the optimized feature parameters.
[0015] In addition, to achieve the above-mentioned purpose, the present application also provides a device for acquiring myocardial infarction prediction data based on a simulated annealing algorithm. The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is configured to implement the steps of a method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm as provided in any of the above-mentioned technical solutions.
[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the method for acquiring myocardial infarction prediction data based on the simulated annealing algorithm as provided in any of the above-mentioned technical solutions are implemented.
[0017] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the method for acquiring myocardial infarction prediction data based on the simulated annealing algorithm as provided in any of the above-mentioned technical solutions.
[0018] One or more technical solutions proposed in this application have at least the following technical effects: By constructing multiple feature parameters for myocardial infarction prediction, these feature parameters are strongly correlated with the myocardial infarction prediction features for predicting myocardial infarction, and the myocardial infarction prediction features are physiological features related to the onset of myocardial infarction, especially in the early stage of myocardial infarction. Each myocardial infarction prediction feature can correspond to multiple feature parameters. In this way, the myocardial infarction prediction model can be established by combining the above multiple feature parameters with the corresponding myocardial infarction prediction features and weight factors. In the myocardial infarction prediction model, the weight factor is the weight contribution item of each myocardial infarction prediction feature to the acquisition of myocardial infarction prediction data. The myocardial infarction prediction feature combined with the user's feature parameters can accurately predict and judge whether the user has a risk of myocardial infarction in detail; then, the myocardial infarction prediction data is obtained from the physical sign data of a large number of myocardial infarction patients by random sampling, and the data set of the above multiple myocardial infarction prediction features is generated, thereby reducing the specialization and limitation of the feature data caused by various other factors. Using the replica exchange simulated annealing algorithm, combined with the above myocardial infarction prediction model and the data set of the above multiple myocardial infarction prediction features, the myocardial infarction prediction features are input into the myocardial infarction prediction model, and the model results corresponding to the multiple feature parameters can be solved. Because the simulated annealing algorithm is a random optimization algorithm based on the Monte Carlo iterative solution strategy, it can randomly search for the global optimal solution of the objective function in the solution space, and eventually tend to the global optimum, so it can solve and obtain the best optimized feature parameters among the above-mentioned multiple feature parameters. Finally, the above-mentioned optimized feature parameters are substituted into the myocardial infarction prediction model to obtain the optimized myocardial infarction prediction model. The myocardial infarction prediction model is used to predict the results of the myocardial infarction prediction features to be tested, and accurate myocardial infarction prediction data can be obtained. Because the myocardial infarction prediction features are features that are strongly correlated with the onset of myocardial infarction selected from a variety of physiological characteristics of the user, and the replica exchange simulated annealing algorithm can select the best optimized feature parameters for predicting myocardial infarction from multiple feature parameters, the myocardial infarction prediction data acquisition scheme provided in this application can make timely and accurate predictions of the onset of myocardial infarction, avoiding the problem of inaccurate prediction of myocardial infarction data in the prior art, resulting in misjudgment. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0021] Figure 1 A schematic diagram of a flow chart of a first method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm provided in an embodiment of the present application; Figure 2 For this application Figure 1 A schematic diagram of a process for constructing multiple characteristic parameters is provided; Figure 3 For this application Figure 1 A schematic flow chart of a method for establishing a myocardial infarction prediction model provided in the illustrated embodiment; Figure 4 A flowchart of a method for selecting features for predicting myocardial infarction provided in an embodiment of the present application; Figure 5 for Figure 1 A schematic flow chart of a method for solving multiple characteristic parameters provided by the illustrated embodiment; Figure 6 yes Figure 1 A schematic flow chart of a method for acquiring myocardial infarction prediction data provided by the illustrated embodiment; Figure 7 It is a flowchart of a scoring method for a myocardial infarction prediction model provided in an embodiment of the present application; Figure 8 is a parameter-characteristic curve diagram provided in an embodiment of the present application; Fig. 9 It is a structural schematic diagram of a parameter optimization framework of a simulated annealing algorithm provided in an embodiment of the present application; Fig.10 It is a structural schematic diagram of a myocardial infarction prediction data acquisition device based on a simulated annealing algorithm provided in an embodiment of the present application.
[0022] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0023] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.
[0024] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.
[0025] The main solution of the embodiment of the present application is: to establish a myocardial infarction prediction model using multiple feature parameters, myocardial infarction prediction features and weight factors, and to use a simulated annealing algorithm based on replica exchange to solve the optimal multiple feature parameters, i.e., the above-mentioned optimized feature parameters. The simulated annealing algorithm is a random optimization algorithm based on the Monte Carlo iterative solution strategy, which can randomly find the global optimal solution of the objective function in the solution space, and eventually tend to the global optimum, so that the optimal multiple feature parameters can be solved. Finally, the above-mentioned optimal feature parameters are substituted into the myocardial infarction prediction model to obtain the optimized myocardial infarction prediction model, and the myocardial infarction prediction model can be used to accurately obtain myocardial infarction prediction data for users. Because the above-mentioned multiple myocardial infarction prediction features are myocardial infarction prediction features, which are physiological features related to the onset of myocardial infarction, especially the early stage of myocardial infarction, and the replica exchange simulated annealing algorithm can select the feature parameters closest to the symptoms of the early stage of myocardial infarction, i.e., the above-mentioned optimized feature parameters, the myocardial infarction prediction data acquisition scheme provided by the present application can timely and accurately predict the myocardial infarction of patients.
[0026] The existing technology has the following technical problems: the existing myocardial infarction detection methods mostly focus on the in-process diagnosis of myocardial infarction in patients, and it is difficult to accurately predict myocardial infarction in patients in the early stages of the disease or when no obvious signs appear. Send patients to the doctor in time, which easily leads to patients having difficulty in obtaining effective rescue in time and it is difficult to reduce the mortality rate of myocardial infarction.
[0027] To solve the above technical problems, the present application provides a solution, which includes constructing multiple feature parameters for myocardial infarction prediction; establishing a myocardial infarction prediction model based on the multiple feature parameters in combination with corresponding myocardial infarction prediction features and weight factors; generating data sets of multiple myocardial infarction prediction features respectively according to the random sampling method; using a replica exchange simulated annealing algorithm, combining the myocardial infarction prediction model and the data sets of multiple myocardial infarction prediction features to obtain the optimal multiple feature parameters; using the optimal multiple feature parameters to optimize the myocardial infarction prediction model, and using the optimized myocardial infarction prediction model to obtain myocardial infarction prediction data for users.
[0028] In summary, the myocardial infarction prediction data acquisition scheme based on the simulated annealing algorithm constructed in this application can comprehensively consider a variety of myocardial infarction prediction features that are strongly correlated with the onset of myocardial infarction. This feature can predict the signs of myocardial infarction, and select feature parameters related to the prediction features to further enrich the myocardial infarction prediction model. And by using the replica exchange simulated annealing algorithm to solve the myocardial infarction prediction model, it is possible to select optimized feature parameters, that is, the prediction parameter values that are closest to the early symptoms of myocardial infarction, thereby constructing an accurate myocardial infarction prediction model, which can timely and accurately predict the patient's myocardial infarction condition.
[0029] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device that can realize the above functions, a computer-readable storage medium, etc.
[0030] In order to better understand the above technical solution, the exemplary embodiments of the present application will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments shown here. On the contrary, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.
[0031] To achieve the above technical objectives, see Figure 1 , Figure 1 A method for obtaining myocardial infarction prediction data based on a simulated annealing algorithm is provided for an embodiment of the present application. To achieve the above-mentioned purpose, the method for obtaining myocardial infarction prediction data based on a simulated annealing algorithm includes: S110: Construct multiple characteristic parameters for myocardial infarction prediction. The multiple characteristic parameters correspond to myocardial infarction prediction features, which include but are not limited to chest pain intensity, chest tightness, myocardial infarction ST segment elevation amplitude, and arrhythmia degree. These myocardial infarction prediction features can be screened from the physical sign data of myocardial infarction patients.
[0032] Each MI prediction feature will select corresponding feature parameters, including but not limited to: genetic disease gene carrier status related to MI, myocardial cell ischemia and hypoxia status, pulmonary hypertension status, user's dietary salt and oil intake, user's height, weight and physical condition, user's surgical history and family medical history, etc. These feature parameters can be obtained through big data screening. In this way, for each MI prediction feature, multiple corresponding feature parameters can be selected to construct a prediction parameter set.
[0033] Specifically, as a preferred embodiment, Figure 2 As shown, in the above-mentioned method for obtaining myocardial infarction prediction data, step S110: the step of constructing multiple characteristic parameters for myocardial infarction prediction specifically includes: S111: Acquire a data feature set of multiple myocardial infarction prediction features, wherein the data feature set includes the myocardial infarction prediction features and multiple associated feature parameters.
[0034] For each MI prediction feature, there are multiple associated feature parameters. MI prediction features are features related to the onset of MI, especially features related to the early stage of MI and the asymptomatic period before MI. By selecting these MI prediction features, the user's MI risk can be detected promptly and quickly at the early stage of MI or even in the asymptomatic period before MI.
[0035] S112: For each myocardial infarction prediction feature, characteristic parameters and parameter-characteristic curves of the corresponding myocardial infarction prediction features are respectively established by a fixed variable method.
[0036] Because there are multiple characteristic parameters associated with each myocardial infarction prediction feature, it is only necessary to obtain the top N items (for example, the top 3 items) that are strongly correlated with the myocardial infarction prediction feature. In addition, in order to match the established myocardial infarction prediction model, the parameter-characteristic curve diagram here can include a curve diagram with a characteristic parameter as the independent variable and the myocardial infarction prediction feature as the dependent variable after fixing the values of other characteristic parameters; or a curve diagram with the myocardial infarction prediction feature as a fixed value, other characteristic parameters fixed, the characteristic parameter as the independent variable, and the characteristic energy corresponding to the myocardial infarction prediction feature as the dependent variable. Make the selection based on the actual situation. See Figure 8 , Figure 8 This is a diagram of cardiac energy release corresponding to the T wave under different IU / L conditions of creatine kinase isoenzymes. Under normal circumstances, the normal value of creatine kinase isoenzymes is around 25IU / L.
[0037] S113: curve fitting is performed on the parameter-feature curve graph to obtain the correlation between each feature parameter and the corresponding myocardial infarction prediction feature. Here, by using the curve fitting method, a fitting equation related to each feature parameter and the myocardial infarction prediction feature can be obtained, and then the contribution degree of each feature parameter to the myocardial infarction prediction feature, that is, the correlation with the myocardial infarction prediction feature, is obtained, and then the feature parameters with strong correlation with the myocardial infarction prediction feature are screened.
[0038] S114: In order of correlation from high to low, select the top N feature parameters that are strongly correlated with the myocardial infarction prediction features and set them as correlation parameters. The number of the top N feature parameters here is not fixed, and needs to be determined based on the contribution of the feature parameters to the myocardial infarction prediction features. Usually, a contribution of 80% to the myocardial infarction prediction features is required, and then the corresponding number of feature parameters can be selected. For example, if the user's myocardial infarction-related genes, family genetic disease history, and surgical history can contribute 80% to the user's arrhythmia, then these three parameters are selected as key parameters.
[0039] S115: Use the associated parameters to construct multiple feature parameters. The first N feature parameters corresponding to each myocardial infarction prediction feature are stored together as a data set. When solving the myocardial infarction prediction model, the first N feature parameters are input into the myocardial infarction prediction model to calculate the model result.
[0040] In the technical solution provided by the embodiment of the present application, for each myocardial infarction prediction feature, a parameter-feature curve diagram of the feature parameter and the myocardial infarction prediction feature is established by the fixed variable method; then, the parameter-feature curve diagram is curve-fitted to obtain the correlation of each feature parameter to the myocardial infarction prediction feature, which reflects the contribution of the parameter to the myocardial infarction prediction feature, so that the top N feature parameters that are strongly correlated with the myocardial infarction prediction feature can be selected, and such processing is performed for each myocardial infarction prediction feature, so that the above-mentioned multiple feature parameters can be constructed. The multiple feature parameters include a data set of the top N feature parameters corresponding to each myocardial infarction prediction feature.
[0041] Figure 1 The method for obtaining myocardial infarction prediction data based on the simulated annealing algorithm provided in the illustrated embodiment, after constructing multiple characteristic parameters for myocardial infarction prediction, further includes: S120: Establish a myocardial infarction prediction model based on multiple feature parameters combined with corresponding myocardial infarction prediction features and weight factors. Each myocardial infarction prediction feature has a different prediction weight for myocardial infarction. It is possible that severe chest pain occurs in the early stage of myocardial infarction, but the heart rate is within the normal range. At this time, it is necessary to set the weight factor corresponding to each myocardial infarction prediction feature according to the actual situation, where the sum of all weight factors is 1. In addition, the technical idea of the embodiment of the present application is to use the myocardial infarction prediction model as an energy contribution item for each myocardial infarction prediction feature. During or before the onset of myocardial infarction, the abnormal performance of the heart is actually a manifestation of self-rescue; among them, different myocardial infarction prediction features correspond to the specific behavioral performance of the heart and heart-related tissues and organs, such as abnormal heart rate, increased heartbeat intensity, and elevated ST segment of myocardial infarction, which are actually the performance of the work of the heart and related organs, and will generate corresponding energy. The technical idea of the present application is to split the total energy released by the heart during myocardial infarction into energy values corresponding to multiple myocardial infarction prediction features, thereby constructing a myocardial infarction prediction model.
[0042] Specifically, as a preferred embodiment, Figure 3 As shown, in the above-mentioned method for obtaining myocardial infarction prediction data, step S120 is a step of establishing a myocardial infarction prediction model according to a plurality of characteristic parameters combined with corresponding myocardial infarction prediction features and weight factors, including: S121: Use multiple feature parameters, combined with corresponding myocardial infarction prediction features and weight factors, to construct an expression for myocardial infarction prediction model:
[0043] in, Represents the total myocardial infarction energy, represents the ground state energy, represents the feature energy corresponding to the i-th myocardial infarction prediction feature, represents the j-th weight factor, represents the i-th characteristic parameter, Represents the jth myocardial infarction prediction feature. Different myocardial infarction prediction features have different influences on myocardial infarction, so it is necessary to allocate weights for different myocardial infarction prediction features. Specifically, the weight factor value range can be allocated through actual data analysis.
[0044] It should be noted that It represents the total myocardial infarction energy, which can be expressed by the amount of work done by the heart, which can be obtained by considering the work done by the heart each time it beats; the calculation formula of the heart's work per stroke, taking the left ventricle pumping blood as an example, for example: stroke work = stroke volume × (left ventricular internal pressure during ejection phase - left ventricular end-diastolic pressure), and then multiplying it by the number of heart beats; of course, it can also refer to the pulsation of the coronary artery. Similarly, the ground state energy Indicates the amount of work done by the user's heart under normal circumstances. It represents the characteristic energy corresponding to the j-th myocardial infarction prediction feature, and these energies can be calculated through the above-mentioned myocardial infarction prediction features and the characteristic parameters corresponding to the myocardial infarction prediction features; specifically, refer to the above-mentioned curve graph with the characteristic parameters as the independent variable and the characteristic energy corresponding to the myocardial infarction prediction feature as the dependent variable, and the curve equation obtained by fitting.
[0045] S122: Constructing a loss function of the myocardial infarction prediction model based on the energy difference between the model calculated value and the actual value of the myocardial infarction energy.
[0046] Here, it is assumed that the model calculation value of the myocardial infarction energy corresponding to the i-th characteristic parameter is calculated by the expression of the myocardial infarction prediction model in step S121 above: The actual energy of myocardial infarction is calculated by the above formula. ; The loss function of constructing the myocardial infarction prediction model is as follows:
[0047] in, It represents the energy difference between the actual value of myocardial infarction energy and the value calculated by the model.
[0048] From the loss function of the above myocardial infarction prediction model, it can be seen that the simulated annealing algorithm will calculate the total myocardial infarction energy for each characteristic parameter i. When the difference between the model calculated value and the actual value of the myocardial infarction energy is the smallest, it means that the characteristic parameter of the myocardial infarction energy is optimal at this time.
[0049] in addition, Figure 1 The method for acquiring myocardial infarction prediction data based on the simulated annealing algorithm provided in the illustrated embodiment further includes the following steps after establishing the myocardial infarction prediction model: S130: From the physical sign data of myocardial infarction patients, a data set of multiple myocardial infarction prediction features is obtained by screening according to the random sampling method. In the present application, the physical sign data of myocardial infarction patients can be obtained by big data sampling, and then the data corresponding to the myocardial infarction prediction features are found, and finally the random sampling method is used to select the data of each myocardial infarction prediction feature at random intervals, so that the data set of the above-mentioned multiple myocardial infarction prediction features can be obtained. Among them, the use of random sampling can reduce the influence of the specialization and limitation of feature data caused by various other factors.
[0050] Figure 1 The method for obtaining myocardial infarction prediction data based on the simulated annealing algorithm provided in the illustrated embodiment further includes the following steps after screening and obtaining a data set of multiple myocardial infarction prediction features: S140: Select feature data from the data set and input it into the myocardial infarction prediction model, and use the replica exchange simulated annealing algorithm to solve and obtain optimized feature parameters. The feature data here refers to the feature data of myocardial infarction prediction in the above data set. The replica exchange simulated annealing algorithm, referred to as RESA, is a variant of the simulated annealing algorithm, which is mainly used to solve complex optimization problems, especially those with multiple local optimal solutions. In the traditional simulated annealing algorithm, the system starts from an initial state and performs random search at a series of temperatures, avoiding falling into the local optimum by accepting inferior solutions with a certain probability. RESA introduces the concept of "replicas", and further improves the ability of global search by exchanging states between multiple replicas at different temperatures. The replica contains multiple sets of feature parameters, each of which is a set of the above-mentioned multiple feature parameters. Using the replica exchange simulated annealing algorithm, after substituting the data of multiple myocardial infarction prediction features into the myocardial infarction prediction model, the feature parameters can be valued within the value range of the feature parameters to obtain a set of values of different feature parameters, and the total myocardial infarction energy obtained by the myocardial infarction prediction model is used to solve the optimal multiple feature parameters among the above-mentioned multiple feature parameters, that is, the optimized feature parameters. The optimized feature parameters select the best parameters in a series of parameter samples, which can optimize the prediction performance of the myocardial infarction prediction data acquisition method and its equipment, and make the related equipment universal for most patients. Because it is difficult to obtain all of the patient's multiple characteristic parameters, the use of the above-mentioned optimized characteristic parameters can universally predict myocardial infarction data.
[0051] Specifically, as a preferred embodiment, Figure 4 As shown, in the above-mentioned method for obtaining myocardial infarction prediction data, step S140: the step of selecting feature data in the data set and inputting it into the myocardial infarction prediction model includes: S141: Convert the data set of multiple myocardial infarction prediction features into a multidimensional prediction feature graph. In the embodiment of the present application, the dimension in the multidimensional prediction feature graph corresponds to the number of types of myocardial infarction prediction features. Because each of the multiple myocardial infarction prediction features includes a data set, the multidimensional prediction feature graph is actually a data combination of different types of myocardial infarction prediction features.
[0052] S142: Randomly select any position coordinate point from the multi-dimensional prediction feature map. The position coordinate point here does not necessarily contain the myocardial infarction prediction feature. The present application establishes a coordinate system for the multi-dimensional prediction feature map, so any point in the map can have a position coordinate.
[0053] S143: Using a K-nearest neighbor algorithm, a set of myocardial infarction prediction feature sets within a predetermined distance range is selected with the position coordinate point as the center, wherein the set of myocardial infarction prediction feature sets includes a plurality of different types of myocardial infarction prediction features.
[0054] The so-called K nearest neighbor algorithm is to give a training data set, and for a new input instance, find the K instances closest to the instance in the training data set. Most of these K instances belong to a certain class, and the input instance is classified into this class. Since the set of abnormal myocardial infarction features input into the myocardial infarction prediction model at one time in the present application specification is allocated by group, a set of myocardial infarction prediction feature sets includes multiple different types of myocardial infarction prediction features, so that the values of multiple different types of myocardial infarction prediction features can be input into the myocardial infarction prediction model for solution. Similarly, because the K nearest neighbor algorithm selects feature data within the neighbor range, different types of myocardial infarction prediction feature data will not differ greatly. This is because different types of myocardial infarction prediction features are interrelated, and there will be no huge differences in the values of different myocardial infarction prediction features when a myocardial infarction occurs. For example, when a myocardial infarction occurs, when the ST segment is elevated, there will be no stable heart rhythm.
[0055] S144: Inputting a plurality of different types of myocardial infarction prediction features into a myocardial infarction prediction model for solving.
[0056] After selecting a set of myocardial infarction prediction feature sets, different types of myocardial infarction prediction features in the set are input into the myocardial infarction prediction model, and the model results of the myocardial infarction prediction model can be calculated in combination with the above-mentioned feature parameters.
[0057] Finally, randomly select any position point from the multi-dimensional prediction feature map and the subsequent steps are re-executed until the multi-dimensional prediction feature map is traversed.
[0058] Specifically, as a preferred embodiment, Figure 5As shown, in the above-mentioned method for obtaining myocardial infarction prediction data, step S140: using a replica exchange simulated annealing algorithm, combining a myocardial infarction prediction model and a data set of multiple myocardial infarction prediction features, and solving the step of obtaining the optimal multiple feature parameters includes: S145: Configure the parameter optimization framework corresponding to the replica exchange simulated annealing algorithm, combined with Fig. 9 It can be seen from the parameter optimization framework shown that the parameter optimization framework includes a recursive loop 100, an exchange loop 200 and a scanning loop 300. The recursive loop 100 is used to use multiple parameter copies to substitute them into the myocardial infarction prediction model in parallel for calculation. At the end of each recursive loop 100, any one or more groups of characteristic parameters in the parameter copy are replaced to simulate different characteristic parameters within the parameter value range, and the loss function value is calculated once at the end of each recursive loop to minimize the loss function. The exchange loop 200 is used to exchange the characteristic parameters in two parameter copies, which prevents the model from falling into a local optimal value.
[0059] S146: Generate multiple parameter copies, where each parameter copy represents multiple groups of characteristic parameters, and each group of characteristic parameters is a collection of multiple characteristic parameters. In the above multiple groups of characteristic parameters, each characteristic parameter has a certain value range, which can be set by relevant personnel based on clinical experience, so that the value ranges of multiple characteristic parameters can form a parameter distribution area, and multiple groups of characteristic parameters can be obtained by using parameter copies to select in the parameter distribution area. As long as the position of the parameter copy in the parameter distribution area is moved, the entire parameter distribution area can be covered, and the characteristic parameters can be exhaustively enumerated.
[0060] S147: In the recursive loop, multiple parameter copies are used in parallel to execute the simulated annealing algorithm. The parameter copies include multiple sets of characteristic parameters, and the multiple parameter copies execute the simulated annealing algorithm in parallel, so that the parameter exhaustive efficiency of the simulated annealing algorithm can be improved.
[0061] S148: In the exchange cycle, the characteristic parameters in two adjacent parameter copies are exchanged according to a predetermined temperature exchange strategy. The values of the characteristic parameters in the two adjacent parameter copies are similar, so by exchanging the parameters in the adjacent copies in the exchange cycle, the model can be prevented from falling into the local minimum value of the characteristic parameter, and the local minimum value is used as the global optimal value.
[0062] S149: In the scanning cycle, multiple characteristic parameters and characteristic data are used to execute the simulated annealing algorithm, and the energy difference of the myocardial infarction prediction model is calculated using the loss function. The loss function is the energy difference between the actual value of the energy released by the myocardial infarction under specific characteristic parameters and the model calculated value. According to the above myocardial infarction prediction model, the smaller the energy difference, the closer the model calculated value is to the actual value, and the closer the parameter is to the optimal value.
[0063] S1410: Filter a group of characteristic parameters corresponding to the minimum energy difference as the optimal multiple characteristic parameters. Each group of characteristic parameters includes a combination of data values of the multiple characteristic parameters.
[0064] The technical solution provided in the embodiment of the present application uses a replica exchange simulated annealing method to optimize the feature parameters through recursive loops, exchange loops and scanning loops, which can efficiently fit the feature parameters and ensure that the feature parameters are the global optimal solution. The obtained feature parameters take into account the comprehensive impact of the interaction of various myocardial infarction prediction features.
[0065] Figure 1 The method for obtaining myocardial infarction prediction data based on the simulated annealing algorithm provided in the illustrated embodiment, after solving and obtaining the optimal multiple characteristic parameters, further includes: S150: Use the myocardial infarction prediction model containing optimized characteristic parameters to predict the results of the myocardial infarction prediction features to be tested, and obtain myocardial infarction prediction data. The myocardial infarction prediction data corresponding to the optimal multiple characteristic parameters are closest to the true value. Therefore, the myocardial infarction prediction model that substitutes the above optimized characteristic parameters is the most accurate in predicting myocardial infarction. Using the optimized myocardial infarction prediction model for prediction can timely and accurately predict the user's myocardial infarction condition and improve the user's survival rate. Among them, the myocardial infarction prediction data includes the total myocardial infarction energy and the degree of danger of myocardial infarction caused by the above prediction.
[0066] Specifically, as a preferred embodiment, Figure 6 As shown, in the above-mentioned method for obtaining myocardial infarction prediction data, step S150: using the myocardial infarction prediction model including the optimized feature parameters to predict the result of the myocardial infarction prediction feature to be tested, and the step of obtaining myocardial infarction prediction data includes: S151: Real-time monitoring of the user's various myocardial infarction prediction characteristics. The user's myocardial infarction prediction characteristics include but are not limited to chest pain intensity, chest tightness, myocardial infarction ST segment elevation amplitude and arrhythmia degree, which can be measured by diagnostic equipment.
[0067] S152: Input the above-mentioned multiple myocardial infarction prediction features into a myocardial infarction prediction model to predict and obtain myocardial infarction prediction data.
[0068] S153: According to the risk level of the myocardial infarction prediction data, select the corresponding warning signal and presentation method, and send the myocardial infarction prediction data.
[0069] The myocardial infarction prediction data is presented to specialized treatment personnel as the intermediate value of myocardial infarction judgment as one of the basis for disease judgment. For example, according to the prediction results, choose the presentation method, such as outputting a red warning sign for high risk and a yellow warning sign for medium risk. Do not use it directly as a diagnosis result. When the myocardial infarction prediction model predicts that the user is at risk of myocardial infarction, a myocardial infarction warning signal is sent remotely. The ways of sending the myocardial infarction signal remotely include but are not limited to Bluetooth, WIFI, mobile network, zigbeeNFC and NearLink. When the myocardial infarction prediction model predicts that the user is at risk of myocardial infarction, the myocardial infarction warning signal is sent remotely, which can prompt the relevant treatment personnel to rescue the user in a timely and effective manner.
[0070] In addition, as a preferred embodiment, Figure 7 As shown, the above-mentioned method for obtaining myocardial infarction prediction data, after the step S150: using the optimized myocardial infarction prediction model to obtain myocardial infarction prediction data for the user, further comprises: S201: Counting the model prediction value of the myocardial infarction prediction model within a predetermined time.
[0071] S202: Calculate the Euclidean distance between the model prediction value and the actual value of myocardial infarction.
[0072] Euclidean distance refers to the distance between two points in two-dimensional space or three-dimensional space. The formula for calculating Euclidean distance in two-dimensional space is as follows: d =SPRT[ (x1-x2)^2 + (y1-y2)^2 ] Although the model prediction value and the myocardial infarction actual value are both scalars in the embodiments of the present application, the model prediction value is obtained by using the optimal feature parameters in the implementation of the present application. Therefore, the actual difference between the model prediction value and the myocardial infarction actual value is very small. Therefore, the present application uses the Euclidean distance algorithm to calculate the difference between the model prediction value and the myocardial infarction actual value. The smaller the Euclidean distance value between the two, the higher the similarity between the model prediction value and the myocardial infarction actual value. Therefore, by scoring the myocardial infarction prediction model according to the similarity, an accurate evaluation of the myocardial infarction prediction model can be obtained.
[0073] S203: Use the Euclidean distance to calculate the similarity between the model prediction value and the actual value of myocardial infarction, and score the myocardial infarction prediction model according to the similarity. The similarity here can be obtained by using the ratio between the Euclidean distance calculated above and the actual value of myocardial infarction. By setting the ratio range, the accurate score of the myocardial infarction prediction model can be obtained. Specifically, if the ratio is between 0-5%, the score is 9 points; if the ratio is between 5%-8%, the score is 8 points.
[0074] S204: When the score of the myocardial infarction prediction model is less than or equal to a predetermined score threshold, the replica exchange simulated annealing algorithm is reused to solve the optimized feature parameters. A predetermined score threshold is defined here. If the score of the myocardial infarction prediction model is less than or equal to the threshold, it means that the prediction accuracy of the myocardial infarction prediction model is not high enough. At this time, the replica exchange simulated annealing algorithm is reused to solve the optimal feature parameters on the flower-spreading girl, thereby obtaining a myocardial infarction prediction model with higher prediction accuracy.
[0075] In summary, the technical solution provided by the above embodiment of the present application constructs multiple feature parameters for myocardial infarction prediction, and these feature parameters are strongly correlated with the myocardial infarction prediction features for predicting myocardial infarction, and the myocardial infarction prediction features are physiological features related to the onset of myocardial infarction, especially the early stage of myocardial infarction. Each myocardial infarction prediction feature can correspond to multiple feature parameters. In this way, the above multiple feature parameters are combined with the corresponding myocardial infarction prediction features and weight factors to establish a myocardial infarction prediction model. In the myocardial infarction prediction model, the weight factor is the weight contribution item of each myocardial infarction prediction feature to the acquisition of myocardial infarction prediction data. The myocardial infarction prediction feature is combined with the user's feature parameters to accurately predict and judge whether the user has a risk of myocardial infarction in detail. Then, the myocardial infarction prediction data is obtained from the big data of the myocardial infarction prediction feature by random sampling, and the data set of the above multiple myocardial infarction prediction features is generated, thereby reducing the influence of feature specialization and localization caused by various other factors. Using the replica exchange simulated annealing algorithm, combined with the above myocardial infarction prediction model and the data set of the above multiple myocardial infarction prediction features, the myocardial infarction prediction feature is input into the myocardial infarction prediction model, and the model results corresponding to the multiple feature parameters can be solved. Because the simulated annealing algorithm is a random optimization algorithm based on the Monte Carlo iterative solution strategy, it can randomly search for the global optimal solution of the objective function in the solution space, and eventually tend to the global optimum. Therefore, it is possible to solve and obtain the optimal multiple feature parameters. Finally, the above-mentioned optimized feature parameters are substituted into the myocardial infarction prediction model to obtain the optimized myocardial infarction prediction model. The myocardial infarction prediction model is used to predict the results of the myocardial infarction prediction features to be tested, and accurate myocardial infarction prediction data can be obtained. Because the myocardial infarction prediction features are features that are strongly correlated with the onset of myocardial infarction selected from the user's physiological characteristics, and the replica exchange simulated annealing algorithm can select the optimal optimized feature parameters for predicting myocardial infarction from multiple feature parameters, the myocardial infarction prediction data acquisition scheme provided in this application can make timely and accurate predictions of the onset of myocardial infarction, avoiding the problem of inaccurate prediction of myocardial infarction data in the prior art, which causes misjudgment.
[0076] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the model building method of the magnetic system of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.
[0077] In addition, the myocardial infarction prediction data acquisition device based on simulated annealing algorithm provided in the present application includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the myocardial infarction prediction data acquisition method of any of the above-mentioned embodiments.
[0078] Reference below Fig.10, which shows a schematic diagram of the structure of a myocardial infarction prediction data acquisition device based on a simulated annealing algorithm suitable for implementing the embodiment of the present application. The myocardial infarction prediction data acquisition device in the embodiment of the present application can include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.10 The myocardial infarction prediction data acquisition device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.
[0079] like Fig.10 As shown, the myocardial infarction prediction data acquisition device based on the simulated annealing algorithm can include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the myocardial infarction prediction data acquisition device are also stored in RAM1004. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can enable the myocardial infarction prediction data acquisition device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a model building device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.
[0080] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.
[0081] The myocardial infarction prediction data acquisition device based on the simulated annealing algorithm provided by the present application adopts the myocardial infarction prediction data acquisition method in the above embodiment, which can solve the technical problem that the existing technical methods are difficult to timely and accurately predict the myocardial infarction of patients. Compared with the prior art, the beneficial effects of the myocardial infarction prediction data acquisition device provided by the present application are the same as the beneficial effects of the myocardial infarction prediction data acquisition method provided by the above embodiment, and the other technical features of the myocardial infarction prediction data acquisition device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.
[0082] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.
[0083] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
[0084] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm in the above-mentioned embodiment.
[0085] The computer-readable storage medium provided in the present application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.
[0086] The computer-readable storage medium can be included in the model building device of the magnetic system; or it can exist independently without being assembled into the model building device of the magnetic system.
[0087] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the model building device, the myocardial infarction prediction data acquisition device: constructs multiple feature parameters for myocardial infarction prediction; establishes a myocardial infarction prediction model based on the multiple feature parameters combined with corresponding myocardial infarction prediction features and weight factors; generates data sets of multiple myocardial infarction prediction features respectively according to the random sampling method; uses a replica exchange simulated annealing algorithm to combine the myocardial infarction prediction model and the data sets of multiple myocardial infarction prediction features to solve and obtain the optimal multiple feature parameters; uses the optimal multiple feature parameters to optimize the myocardial infarction prediction model, and uses the optimized myocardial infarction prediction model to acquire myocardial infarction prediction data for users.
[0088] Computer program code for performing the operations of the present application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).
[0089] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes also be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs a specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.
[0090] The modules involved in the embodiments of the present application can be implemented by software or hardware, wherein the name of the module does not limit the unit itself in some cases.
[0091] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned method for obtaining myocardial infarction prediction data based on the simulated annealing algorithm, and can solve the technical problem that the prior art is difficult to predict the patient's myocardial infarction in a timely and accurate manner. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the method for obtaining myocardial infarction prediction data based on the simulated annealing algorithm provided in the above-mentioned embodiment, and will not be repeated here.
[0092] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method for acquiring myocardial infarction prediction data based on the simulated annealing algorithm as described above.
[0093] The computer program product provided by this application can solve the technical problem that the existing methods are difficult to predict the myocardial infarction of patients in a timely and accurate manner. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the myocardial infarction prediction data acquisition method based on the simulated annealing algorithm provided in the above embodiment, which will not be repeated here.
[0094] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.
Claims
1. A method for acquiring myocardial infarction prediction data based on simulated annealing algorithm, characterized in that: include: Construct multiple characteristic parameters for myocardial infarction prediction; Establishing a myocardial infarction prediction model according to the plurality of characteristic parameters combined with corresponding myocardial infarction prediction features and weight factors; From the physical sign data of myocardial infarction patients, a data set of various myocardial infarction prediction features was obtained by screening according to the random sampling method; Selecting characteristic data from the data set and inputting it into the myocardial infarction prediction model, and using a replica exchange simulated annealing algorithm to solve and obtain optimized characteristic parameters; The myocardial infarction prediction model including the optimized characteristic parameters is used to predict the results of the myocardial infarction prediction features to be tested, thereby obtaining myocardial infarction prediction data.
2. The method according to claim 1, characterized in that The step of constructing a plurality of characteristic parameters for myocardial infarction prediction includes: Acquire a data feature set of multiple myocardial infarction prediction features, wherein the data feature set includes the myocardial infarction prediction features and multiple associated feature parameters; For each myocardial infarction prediction feature, a parameter-feature curve diagram of the feature parameter and the corresponding myocardial infarction prediction feature is established respectively by a fixed variable method; Performing curve fitting on the parameter-characteristic curve diagram to obtain the correlation between each characteristic parameter and the corresponding myocardial infarction prediction feature; According to the order of the correlation from high to low, the first N feature parameters that are strongly correlated with the myocardial infarction prediction feature are selected and set as correlation parameters; The plurality of characteristic parameters are constructed using the associated parameters.
3. The method according to claim 1 or 2, characterized in that: The step of establishing a myocardial infarction prediction model according to the plurality of characteristic parameters combined with corresponding myocardial infarction prediction features and weight factors comprises: Using the multiple feature parameters, combined with the corresponding myocardial infarction prediction features and weight factors, an expression of a myocardial infarction prediction model is constructed: in, Represents the total myocardial infarction energy, represents the ground state energy, represents the feature energy corresponding to the i-th myocardial infarction prediction feature, Indicates j Item weight factor, Indicates i Item characteristic parameters, Indicates j Myocardial infarction prediction features; According to the energy difference between the model calculated value and the actual value of the myocardial infarction energy, a loss function of the myocardial infarction prediction model is constructed.
4. The method according to claim 3, characterized in that The step of selecting the myocardial infarction prediction feature data in the data set and inputting it into the myocardial infarction prediction model comprises: Converting the data set of the multiple myocardial infarction prediction features into a multidimensional prediction feature map; Randomly select any position coordinate point from the multi-dimensional prediction feature map; Using a K-nearest neighbor algorithm, a set of myocardial infarction prediction feature sets within a predetermined distance range is selected with the position coordinate point as the center, wherein the set of myocardial infarction prediction feature sets includes the multiple myocardial infarction prediction features; Inputting the multiple myocardial infarction prediction features into the myocardial infarction prediction model for solving; Re-execute the random selection of any position point from the multi-dimensional prediction feature map and subsequent steps until the multi-dimensional prediction feature map is completely traversed.
5. The method according to claim 3, characterized in that: The step of using the replica exchange simulated annealing algorithm to solve and obtain the optimized characteristic parameters includes: Configuring a parameter optimization framework corresponding to the replica exchange simulated annealing algorithm, wherein the parameter optimization framework includes a recursive loop, an exchange loop, and a scanning loop; Generate multiple parameter copies, wherein each parameter copy represents a plurality of groups of the characteristic parameters, and each group of the characteristic parameters is a data set of a group of the plurality of characteristic parameters; In the recursive loop, executing a simulated annealing algorithm using the multiple parameter copies in parallel; In the exchange cycle, the characteristic parameters in two adjacent parameter copies are exchanged according to a predetermined temperature exchange strategy; In the scanning cycle, the plurality of characteristic parameters and the characteristic data are used to execute a simulated annealing algorithm, and the energy difference of the myocardial infarction prediction model is calculated using the loss function; A set of characteristic parameters corresponding to the minimum energy difference is selected as the optimized characteristic parameters.
6. The method according to claim 1, characterized in that The step of using the myocardial infarction prediction model including the optimized feature parameters to predict the result of the myocardial infarction prediction feature to be tested comprises: Real-time monitoring of multiple myocardial infarction prediction features of the user to be tested; Inputting the multiple myocardial infarction prediction features into the myocardial infarction prediction model to predict and obtain the myocardial infarction prediction data; According to the risk level of the myocardial infarction prediction data, a corresponding warning signal and presentation method are selected, and the myocardial infarction prediction data is sent.
7. The method according to claim 6, characterized in that After the step of using the myocardial infarction prediction model including the optimized feature parameters to predict the result of the myocardial infarction prediction feature to be tested to obtain myocardial infarction prediction data, the method further includes: Counting the model prediction value of the myocardial infarction prediction model within a predetermined time; Calculating the Euclidean distance between the model predicted value and the actual value of myocardial infarction; Using the Euclidean distance to calculate the similarity between the model prediction value and the myocardial infarction actual measured value, and scoring the myocardial infarction prediction model according to the similarity; When the score of the myocardial infarction prediction model is less than or equal to a predetermined score threshold, the replica exchange simulated annealing algorithm is reused to solve the optimized feature parameters.
8. A device for acquiring data for predicting myocardial infarction based on a simulated annealing algorithm, characterized in that: The device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm as described in any one of claims 1 to 7.
9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm as described in any one of claims 1 to 7 are implemented.
10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method for acquiring myocardial infarction prediction data based on a simulated annealing algorithm as claimed in any one of claims 1 to 7 are implemented.
