A method and system for constructing a risk assessment model for myocardial infarction patients

Abstract

The present invention relates to the field of model construction technology, and in particular to a method and system for constructing a risk assessment model for myocardial infarction patients. The method comprises the following steps: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system, and performing electrocardiogram signal-vascular motion mapping processing based on the coronary angiography images of myocardial infarction patients, while simultaneously performing blocked electrocardiogram signal pre-gradient feedback identification to obtain blocked electrocardiogram pre-gradient feedback data; performing electrocardiogram signal phase offset analysis on the blocked electrocardiogram pre-gradient feedback data, and constructing a risk assessment model for myocardial infarction patients to obtain a risk assessment model for myocardial infarction patients; and performing automated monitoring firmware design on the risk assessment model for myocardial infarction patients to obtain monitoring firmware for the risk assessment model for myocardial infarction patients. The present invention makes the assessment model construction technology more perfect by optimizing the construction technology of the risk assessment model for myocardial infarction patients.

Description

Technical Field

[0001] The present invention relates to the technical field of model construction, and in particular to a method and system for constructing a risk assessment model for myocardial infarction patients. Background Art

[0002] Myocardial infarction (MI) is one of the leading causes of death and disability worldwide. Its etiology is complex and diverse, involving multiple risk factors such as hypertension, hyperlipidemia, smoking, and diabetes. Therefore, early risk assessment in patients with MI is crucial for improving prognosis, developing personalized treatment plans, and reducing mortality. Previous risk assessment models, such as the Framingham score and the GRACE score, while widely used in clinical practice, still have limitations in terms of accuracy and applicability. With the rapid development of artificial intelligence and big data technologies, novel MI risk assessment models based on multivariate analysis and machine learning algorithms have been proposed. These models integrate clinical indicators, laboratory data, and imaging features to provide a more comprehensive and dynamic assessment of MI risk. Furthermore, the concept of personalized medicine has promoted the application of models in specific populations, such as elderly patients and those with multiple comorbidities. The development of an effective MI risk assessment model requires considering multidimensional patient data and combining advanced statistical methods and computer algorithms to improve prediction accuracy and clinical applicability, thereby providing clinicians with a powerful decision-making tool. However, the traditional method of constructing a risk assessment model for myocardial infarction patients has problems such as large errors in risk assessment of myocardial infarction patients and inaccurate identification of risk antecedent factors for myocardial infarction patients. Summary of the Invention

[0003] Based on this, it is necessary to provide a method and system for constructing a risk assessment model for myocardial infarction patients to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a method for constructing a risk assessment model for myocardial infarction patients is provided, the method comprising the following steps:

[0005] Step S1: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system to obtain vital sign record data and coronary angiography images of myocardial infarction patients, respectively; constructing an arterial angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients to obtain an arterial angiography multi-view 3D model;

[0006] Step S2: performing ECG signal-vasomotion mapping processing on the multi-view 3D arterial angiography model to obtain ECG signal-vasomotion mapping data; performing vascular occlusion increment calculation on the ECG signal-vasomotion mapping data to obtain vascular occlusion increment data; performing ECG signal pre-gradient feedback identification on the vascular occlusion increment data based on the ECG signal-vasomotion mapping data to obtain ECG pre-gradient feedback data;

[0007] Step S3: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; using a random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the blocked feedback ECG signal phase offset data to obtain a risk assessment model for myocardial infarction patients;

[0008] Step S4: Design an automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the risk assessment model monitoring firmware for myocardial infarction patients, and send the risk assessment model monitoring firmware for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

[0009] The present invention uses a medical information system to acquire vital sign data and coronary angiography images from patients with myocardial infarction (MI), providing comprehensive patient information, including clinical symptoms, signs, and detailed images of vascular structure. This comprehensive data facilitates accurate assessment of the patient's condition. Coronary angiography provides a high-resolution view of the patient's coronary artery health, helping to clearly identify the specific location and extent of vascular stenosis or blockage. By constructing a multi-view three-dimensional model, a stereoscopic view of the coronary arteries can be obtained, providing more information than a two-dimensional image, helping doctors better understand the complex structure and pathological conditions of the vessels. Mapping ECG signals to vascular motion correlates ECG data with vascular motion, allowing observation of how the heart's electrical activity affects vascular motion. This dynamic analysis helps identify vascular responses to ECG signal abnormalities in MI patients, providing additional diagnostic information. By calculating vascular obstruction increments based on ECG signal-vascular motion mapping data, the degree and changes of vascular obstruction can be quantified. Blockage ECG signal pre-gradient feedback can detect potential ECG signal abnormalities in advance, providing early warning before or at the early stages of MI. This facilitates early intervention and reduces the occurrence or severity of MI. Pre-gradient feedback can help improve the accuracy of myocardial infarction (MI) diagnosis, avoid missed opportunities for early diagnosis, and enhance overall patient outcomes. ECG signal phase shift analysis can reveal potential anomalies in ECG signals, particularly during MI, where signal phase changes reflect abnormal cardiac status. This allows for earlier and more accurate detection of MI signals. Obtaining blocked feedback ECG signal phase shift data provides detailed signal characteristics, which helps build more accurate MI risk assessment models. The random forest algorithm can process complex data sets and exploit the nonlinear characteristics of ECG signals to construct a robust MI risk assessment model. This model integrates multiple data features to provide reliable risk assessment results. Random forests have strong predictive power, can identify patients at high risk for MI, and are highly robust to noise and missing data. The design of automated monitoring firmware enables real-time tracking of the risk status of MI patients. This automated system continuously monitors patients' ECG data and provides timely feedback based on the risk assessment model. The design of automated firmware makes the model more efficient and convenient in practical applications, enabling better integration into the healthcare system and improving ease of use and reliability. Sending monitoring firmware to a cloud platform facilitates centralized management and analysis of patient data. The cloud platform can process large amounts of data, conduct real-time risk assessments, and provide timely feedback to medical staff. The cloud platform provides remote access, allowing medical staff to access risk assessment results at any time, facilitating cross-regional and multi-institutional collaboration and decision-making, and improving the accessibility and responsiveness of medical services.Therefore, the present invention is an improvement on the traditional method of constructing a risk assessment model for myocardial infarction patients, which solves the problems of large errors in risk assessment of myocardial infarction patients and inaccurate identification of risk antecedent factors for myocardial infarction patients in the traditional method of constructing a risk assessment model for myocardial infarction patients. It reduces the error in risk assessment of myocardial infarction patients and improves the accuracy of identifying risk antecedent factors for myocardial infarction patients.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system, and obtaining basic data of myocardial infarction patients including vital sign record data and coronary angiography images of myocardial infarction patients;

[0012] Step S12: performing image noise optimization on the coronary angiography image of the myocardial infarction patient to obtain a coronary artery noise optimized image;

[0013] Step S13: performing multi-view reconstruction analysis on the coronary angiography image of the myocardial infarction patient based on the coronary artery noise optimization image to obtain coronary angiography multi-view reconstruction data;

[0014] Step S14: constructing an artery angiography multi-view 3D model based on the coronary angiography multi-view reconstruction data to obtain an artery angiography multi-view 3D model.

[0015] This invention uses a medical information system to collect vital sign data and coronary angiography images, acquiring comprehensive basic data. This includes the patient's clinical signs (such as heart rate and blood pressure) and imaging data, providing foundational and contextual information for subsequent analysis. Accurate collection of vital sign data and coronary angiography images is a prerequisite for ensuring the reliability of subsequent analysis results and helps improve the accuracy of myocardial infarction diagnosis. Image noise optimization reduces noise generated during the imaging process, resulting in clearer images. This improves image quality and facilitates more accurate identification of vascular structure and pathology. Multi-view reconstruction of noise-optimized coronary artery images allows for observation of vascular structure from multiple angles. This method provides a panoramic view of the vessels, facilitating identification of complex vascular lesions and stenosis. Multi-view data helps doctors gain a more comprehensive understanding of the three-dimensional structure of the vessels, improving diagnostic accuracy. The three-dimensional angiography model constructed based on the multi-view reconstruction data provides a stereoscopic view of the vessels, demonstrating their true morphology and pathology better than traditional two-dimensional images. The three-dimensional model intuitively displays the spatial relationships and hemodynamic characteristics of the vessels, facilitating more precise treatment planning.

[0016] Preferably, step S2 includes the following steps:

[0017] Step S21: extracting electrocardiogram records from the myocardial infarction patient's physical sign record data to obtain an electrocardiogram of the myocardial infarction patient;

[0018] Step S22: performing electrocardiogram-vascular motion mapping processing on the multi-view 3D model of arteriography according to the electrocardiogram of the myocardial infarction patient to obtain electrocardiogram-vascular motion mapping data;

[0019] Step S23: using a preset myocardial abnormal behavior recognition model to perform myocardial abnormal behavior recognition on the ECG signal-vasomotion mapping data, to obtain myocardial abnormal behavior data of the myocardial infarction patient;

[0020] Step S24: calculating the vascular occlusion increment on the ECG signal-vascular motion mapping data according to the myocardial abnormal behavior data of the myocardial infarction patient to obtain vascular occlusion increment data;

[0021] Step S25: performing blockage ECG signal pre-gradient feedback identification on the vascular blockage increment data according to the ECG signal-vascular motion mapping data to obtain blockage ECG pre-gradient feedback data.

[0022] The present invention extracts electrocardiogram (ECG) data to obtain detailed information about cardiac electrical activity in patients with myocardial infarction. This information, including key ECG features such as heart rate, rhythm, and ST segment changes, helps understand the electrophysiological state of the heart. The ECG is essential data for assessing the status of myocardial infarction and facilitates accurate signal analysis and identification of abnormal myocardial behavior in subsequent steps. Mapping the ECG signal with a multi-view 3D arterial angiography model combines cardiac electrical activity with vascular motion. This process helps reveal the impact of ECG activity on vascular dynamics, providing a dynamic perspective for further analysis. ECG-vascular motion mapping data allows for a comprehensive consideration of cardiac electrical activity and vascular motion, ultimately understanding the overall cardiovascular status of patients with myocardial infarction. Analyzing the mapping data using a pre-defined abnormal myocardial behavior recognition model can identify abnormal myocardial behavior, such as ischemia and arrhythmias. This facilitates accurate diagnosis of pathological changes in myocardial infarction. Identifying abnormal behavior can provide early warning, enabling doctors to intervene and treat patients promptly, thereby improving patient outcomes. Calculating vascular obstruction increments based on abnormal myocardial behavior data in the ECG-vascular motion mapping data quantifies changes in vascular obstruction. This data helps accurately assess the severity of vascular stenosis or obstruction. Blockage ECG signal pre-gradient feedback allows for proactive analysis of ECG signal abnormalities. This helps identify potential MI risk in advance and provides a basis for timely intervention. Pre-gradient feedback can further optimize risk assessment, reduce the risk of missed or misdiagnosed cases, and improve overall diagnostic accuracy.

[0023] Preferably, step S24 includes the following steps:

[0024] Step S241: calculating abnormal cardiac diastolic frequency based on abnormal myocardial behavior data of myocardial infarction patients to obtain abnormal cardiac diastolic frequency data;

[0025] Step S242: performing blood transport attenuation dynamics simulation on the ECG signal-vascular motion mapping data according to the abnormal diastolic frequency data to obtain blood transport attenuation dynamics simulation data;

[0026] Step S243: performing pressure increment azimuth decomposition calculation on the blood transport attenuation dynamics simulation data to obtain blood pressure increment azimuth decomposition data;

[0027] Step S244: Calculate the vascular occlusion increment based on the blood pressure increment azimuth decomposition data to obtain vascular occlusion increment data.

[0028] The present invention can detect abnormal diastolic behavior by calculating the abnormal diastolic frequency of the heart. This information helps to identify problems of impaired cardiac function at an early stage, especially in patients with myocardial infarction, where abnormal diastolic frequency is an indication of myocardial damage or cardiac insufficiency. Abnormal diastolic frequency data can provide information about cardiac diastolic function, which is very important for diagnosing myocardial lesions or heart failure. Accurate diastolic frequency data helps to assess the overall health of the heart. Based on the abnormal diastolic frequency data, blood transport attenuation dynamic simulation of ECG signal-vascular motion mapping data can simulate the blood transport process in the blood vessels and evaluate the decrease in blood transport efficiency caused by abnormal cardiac function. This helps to understand the hemodynamic characteristics of patients with myocardial infarction. Blood transport attenuation dynamic simulation can reveal the attenuation problem of blood flow, thereby predicting the hemodynamic abnormalities that cause or aggravate myocardial infarction. Pressure increment azimuthal decomposition calculation of blood transport attenuation dynamic simulation data can identify pressure changes at different locations within the blood vessels. This analysis can reveal the pressure changes in different parts of the blood vessels and provide detailed data support for the assessment of vascular lesions. Pressure delta azimuthal decomposition data helps pinpoint the specific location of pressure anomalies, which is crucial for identifying and assessing the exact location and extent of vascular stenosis or obstruction. Calculating vascular obstruction increments based on pressure delta azimuthal decomposition data allows for accurate assessment of vascular obstruction.

[0029] Preferably, performing pressure increment azimuthal decomposition calculation on the blood transport attenuation dynamics simulation data comprises the following steps:

[0030] Performing transport rate attenuation evaluation on the blood transport attenuation dynamics simulation data to obtain blood transport rate attenuation data;

[0031] Performing blood fluidity pressure intensity simulation based on blood transport rate attenuation data to obtain blood fluidity pressure intensity data;

[0032] Performing omnidirectional stress tensor calculation on the blood fluidity pressure intensity data to obtain omnidirectional stress tensor data;

[0033] Based on the omnidirectional stress tensor data, stress spatial distribution fluctuation analysis is performed to obtain stress spatial distribution fluctuation data;

[0034] The spatial potential energy relationship of stress spatial distribution fluctuation data is quantified according to the Bernoulli equation to obtain stress spatial potential energy relationship data;

[0035] The pressure increment azimuth decomposition calculation is performed based on the stress space potential energy relationship data and the stress space distribution fluctuation data to obtain the blood pressure increment azimuth decomposition data.

[0036] The present invention can detect changes in blood flow velocity by evaluating transport velocity attenuation. This assessment reveals blood flow attenuation, particularly in areas of myocardial infarction or vascular stenosis. This is crucial for understanding changes in blood flow. Velocity attenuation data helps predict vascular health and identify areas of vascular obstruction or disease, providing foundational data for further analysis and treatment. Pressure intensity simulation based on blood transport velocity attenuation data can determine the pressure distribution of blood within the blood vessels. This provides important data for understanding blood fluidity and pressure changes within the vessels. This simulation helps identify pressure abnormalities caused by reduced blood fluidity and further assess the functional status of the cardiovascular system. Calculating a full-scale stress tensor based on blood fluidity pressure intensity data allows for a comprehensive analysis of stress distribution within the vessels. The stress tensor provides a comprehensive view of stress in different directions, which is crucial for understanding the stress state of the vessel wall. By calculating the stress tensor, it is possible to assess the burden on the vessel wall and the risk of damage. This is crucial for predicting vascular rupture or other pathological conditions. Spatial stress distribution fluctuation analysis based on the full-scale stress tensor data helps detect stress fluctuations at different spatial locations. This analysis can reveal trends and regional differences in stress within the vessels. Stress fluctuation analysis helps locate areas with vascular problems, such as vascular stenosis and arteriosclerosis. By quantifying the spatial potential energy relationship of stress spatial distribution fluctuation data based on the Bernoulli equation, the energy conversion relationship between stress changes and blood flow can be quantified. This helps understand the relationship between blood flow and pressure changes. Quantifying the spatial potential energy relationship of stress can reveal the patterns of pressure changes and provide a deeper understanding of blood flow and pressure distribution. Performing azimuthal decomposition calculations of pressure increments based on stress spatial potential energy relationship data and stress spatial distribution fluctuation data can provide accurate intravascular pressure increment data. This data helps analyze pressure increments in different parts of the blood vessels. Using pressure increment azimuthal decomposition data, blockages and pressure changes within the blood vessels can be accurately assessed, helping to determine the specific location and severity of the blockage, thereby formulating effective treatment plans.

[0037] Preferably, step S25 includes the following steps:

[0038] Step S251: performing time domain feature analysis on the ECG signal-vascular motion mapping data to obtain ECG signal-vascular motion time domain data;

[0039] Step S252: performing multi-factor correlation analysis on the vascular occlusion increment data based on the ECG signal-vascular motion time domain data to obtain multi-factor correlation data of occlusion;

[0040] Step S253: performing multi-factor cluster analysis on the blocking multi-factor correlation data to obtain blocking multi-factor correlation cluster data;

[0041] Step S254: performing dynamic blockage response simulation on the vascular blockage increment data based on the blockage multi-factor association clustering data to obtain multi-factor dynamic blockage response data;

[0042] Step S255: performing signal response pre-gradient trend identification on the ECG signal-vasomotion mapping data according to the multi-factor dynamic occlusion response data to obtain signal response pre-gradient trend data;

[0043] Step S256: performing blocked ECG signal pre-gradient feedback identification according to the signal response pre-gradient trend data and the multi-factor dynamic blocking response data to obtain blocked ECG signal pre-gradient feedback data.

[0044] The present invention performs time-domain feature analysis on ECG signal-vascular motion mapping data, extracting temporal features and revealing the dynamic patterns of the relationship between ECG signals and vascular motion over time. This analysis helps understand the temporal dependence of cardiac activity and vascular response. Through time-domain features, abnormal patterns in ECG signals and vascular motion can be identified. These abnormalities indicate potential cardiac problems or vascular abnormalities, facilitating early intervention. Multifactor association analysis of vascular obstruction increment data based on ECG signal-vascular motion time-domain data can identify and assess the impact of multiple factors on vascular obstruction. This analysis helps reveal the complex factors and interrelationships that affect vascular obstruction. Multifactor association analysis can provide a more comprehensive understanding of vascular obstruction, helping to identify potential causes and influencing factors, thereby improving diagnostic accuracy. Multifactor cluster analysis of multifactor association data can group data based on similarity, simplifying complex multidimensional data. This clustering helps identify groups of factors with similar characteristics, reducing analysis complexity. Cluster analysis can reveal underlying patterns and relationships in data, helping to understand how different factors contribute to vascular obstruction. Dynamic occlusion response simulation of incremental vascular obstruction data based on multi-factor association clustering data can simulate the dynamic changes in vascular obstruction under different conditions. This helps understand how obstruction changes over time and with varying factors. Dynamic simulation results can help predict occlusion responses, providing a basis for developing personalized treatment strategies and optimizing the timing and approach of interventions. Signal response pre-gradient trend identification of ECG signal-vasomotor mapping data based on multi-factor dynamic occlusion response data can predict future signal trends. This helps to proactively identify abnormal responses and potential problems. Pre-gradient trend data can help establish early warning systems to promptly detect and respond to potential cardiovascular events or changes, thereby improving prevention and intervention effectiveness. Pre-gradient feedback identification of ECG signals based on signal response pre-gradient trend data and multi-factor dynamic occlusion response data can provide information about pre-gradient feedback of ECG signals. This helps to more accurately identify abnormal patterns in ECG signals and their relationship to vascular obstruction.

[0045] Preferably, step S3 includes the following steps:

[0046] Step S31: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data;

[0047] Step S32: performing spatial interpolation processing on the blocked feedback ECG signal phase offset data to obtain signal offset spatial interpolation data;

[0048] Step S33: performing signal compensation on the blocked feedback ECG signal phase offset data according to the signal offset spatial interpolation data to obtain blocked feedback signal phase offset compensation data;

[0049] Step S34: Using the random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the phase offset compensation data of the blocking feedback signal, a risk assessment model for myocardial infarction patients is obtained.

[0050] By analyzing the phase offset of the ECG signal, the present invention can effectively identify interference and noise in the signal and enhance the accuracy of the data. Phase offset analysis helps to discover potential ECG signal anomalies, and spatial interpolation processing can fill in blank areas in the signal due to missing or incomplete data, thereby improving the integrity and continuity of the data. Through spatial interpolation, the spatial resolution of the ECG signal can be improved, making subsequent analysis more accurate. Signal compensation can correct the offset caused by equipment errors or environmental factors and enhance the stability of the signal. The compensated signal is closer to the real data, reducing the error caused by the offset, thereby improving the accuracy of the ECG analysis. The random forest algorithm can process complex data sets, explore the relationship between ECG signals and the risk of myocardial infarction, and provide effective risk assessment. Random forest has high robustness and classification performance, can process high-dimensional data and provide stable prediction results. The algorithm can evaluate the impact of each feature on risk assessment, which helps to understand the key factors of myocardial infarction risk.

[0051] Preferably, step S32 includes the following steps:

[0052] Step S321: calculating the phase offset of the blocked feedback ECG signal phase offset data to obtain the blocked ECG signal phase offset data;

[0053] Step S322: performing nonlinear incremental calculation on the blocked feedback ECG signal phase offset data according to the blocked ECG signal phase offset data to obtain phase offset nonlinear incremental data;

[0054] Step S323: performing spatial increment trend analysis on the phase offset data of the blocked feedback ECG signal according to the phase offset nonlinear increment data to obtain phase offset spatial increment trend data;

[0055] Step S324: performing spatial interpolation processing on the phase offset data of the blocked feedback ECG signal based on the phase offset spatial increment trend data to obtain signal offset spatial interpolation data.

[0056] By calculating the phase offset, the present invention can accurately extract the phase change of the electrocardiogram signal and provide basic data for subsequent analysis. It can identify and quantify the phase offset in the signal, providing the necessary numerical support for subsequent processing and compensation steps. The phase offset data will show nonlinear characteristics, and these complex offset patterns can be better captured through nonlinear incremental calculation. It can provide more accurate compensation data for complex nonlinear offsets, thereby improving the accuracy and stability of the data. By analyzing the spatial increment trend, the potential trends and patterns in the signal can be identified, providing useful information for interpolation processing, and predicting the signal change trend at different locations, thereby optimizing the subsequent spatial interpolation processing and improving the resolution and accuracy of the data. Through spatial interpolation processing, the processed data can be smoothed and the noise and abnormal fluctuations in the signal can be reduced. Spatial interpolation processing can improve the spatial resolution of the signal, making the data more complete and detailed, and facilitating subsequent analysis. It effectively fills the data gaps caused by equipment or environmental factors and improves the continuity and consistency of the overall data.

[0057] Preferably, step S34 includes the following steps:

[0058] Step S341: dividing the blocking feedback signal phase offset compensation data into a data set to obtain a blocking feedback signal phase offset compensation test set and a blocking feedback signal phase offset compensation training set;

[0059] Step S342: performing feedback signal similarity feature selection on the blocking feedback signal phase offset compensation training set to obtain feedback signal similarity feature data;

[0060] Step S343: using a random forest algorithm and performing similar structure incremental learning on the blocking feedback signal phase offset compensation training set based on the similar feature data of the feedback signal to obtain similar structure incremental data;

[0061] Step S344: constructing an initial risk assessment model for myocardial infarction patients based on the similar structure incremental data to obtain an initial risk assessment model for myocardial infarction patients;

[0062] Step S345: performing a model test on the initial myocardial infarction patient risk assessment model based on the blocking feedback signal phase offset compensation test set to obtain a myocardial infarction patient risk assessment model.

[0063] By dividing the data into a training set and a test set, the present invention can effectively train the model and verify its performance to avoid overfitting. The test set is used to evaluate the generalization ability of the model and ensure the prediction accuracy of the model on unseen data. By selecting similar features of the feedback signal, redundant data is reduced and the efficiency of model training is improved. Selecting features with importance and relevance can improve the prediction accuracy and stability of the model. The random forest algorithm can process high-dimensional data and capture complex patterns and relationships in the data through incremental learning of similar structures. Incremental learning allows the model to adapt to new data and improve its ability to process dynamic data. Constructing an initial model provides a basic framework for myocardial infarction risk assessment and can be used for further optimization and verification. The initial model provides a basis for subsequent adjustments and improvements and supports iterative optimization of the model. By verifying the model on the test set, its performance and reliability in real applications are ensured. The testing phase can reveal the shortcomings of the model, provide a basis for improvement, and optimize the actual application effect of the model.

[0064] Preferably, the present invention further provides a system for constructing a risk assessment model for myocardial infarction patients, which is used to execute the method for constructing a risk assessment model for myocardial infarction patients as described above. The system for constructing a risk assessment model for myocardial infarction patients comprises:

[0065] The angiography 3D model construction module is used to collect vital sign data and coronary angiography images of myocardial infarction patients through the medical information system, thereby obtaining vital sign records of myocardial infarction patients and coronary angiography images of myocardial infarction patients, respectively; and construct an angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients, thereby obtaining an angiography multi-view 3D model.

[0066] The ECG signal pre-gradient feedback recognition module is used to perform ECG signal-vascular motion mapping processing on the multi-view 3D model of arteriography to obtain ECG signal-vascular motion mapping data; perform vascular blockage increment calculation on the ECG signal-vascular motion mapping data to obtain vascular blockage increment data; and perform ECG signal pre-gradient feedback recognition on the vascular blockage increment data based on the ECG signal-vascular motion mapping data to obtain ECG pre-gradient feedback data;

[0067] The myocardial infarction patient risk assessment model construction module is used to perform ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; a myocardial infarction patient risk assessment model is constructed using the blocked feedback ECG signal phase offset data using a random forest algorithm to obtain a myocardial infarction patient risk assessment model;

[0068] The automated monitoring operation module is used to design automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the monitoring firmware for the risk assessment model for myocardial infarction patients, and send the monitoring firmware for the risk assessment model for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

[0069] The present invention has the beneficial effect of acquiring myocardial infarction patients' vital sign data and coronary angiography images through a medical information system, providing comprehensive patient information, including clinical symptoms, signs, and detailed images of vascular structure. This comprehensive data facilitates accurate assessment of the patient's condition. Coronary angiography images provide a high-resolution view of the patient's coronary artery health, helping to clearly identify the specific location and extent of vascular stenosis or blockage. By constructing a multi-view three-dimensional model, a stereoscopic view of the coronary arteries can be obtained, providing more information than a two-dimensional image, helping doctors better understand the complex structure and pathological conditions of the vessels. Mapping ECG signals to vascular motion correlates ECG data with vascular motion, allowing observation of how the heart's electrical activity affects vascular motion. This dynamic analysis helps identify vascular responses in patients with myocardial infarction when ECG signals are abnormal, providing additional diagnostic information. By calculating vascular obstruction increments based on ECG signal-vascular motion mapping data, the extent and changes of vascular obstruction can be quantified. Blockage ECG signal pre-gradient feedback recognition can detect potential ECG signal abnormalities in advance, providing early warning before or at the earliest stages of a myocardial infarction. This facilitates early intervention, reducing the occurrence and severity of MI. Pre-gradient feedback can help improve the accuracy of MI diagnosis, avoid missed opportunities for early diagnosis, and enhance overall patient outcomes. ECG signal phase shift analysis can reveal potential anomalies in ECG signals. In particular, during MI, signal phase shifts reflect abnormal cardiac conditions. This allows for earlier and more accurate detection of MI signals. Obtaining blocked feedback ECG signal phase shift data provides detailed signal characteristics, which helps build more accurate MI risk assessment models. The random forest algorithm can process complex data sets and exploit the nonlinear characteristics of ECG signals to construct a robust MI risk assessment model. This model integrates multiple data features to provide reliable risk assessment results. Random forests have strong predictive power, can identify patients at high risk of MI, and are highly robust to noise and missing data. Automated monitoring firmware can be designed to track the risk status of MI patients in real time. This automated system continuously monitors patients' ECG data and provides timely feedback based on the risk assessment model. The design of automated firmware makes the model more efficient and convenient in practical applications, enabling better integration into the healthcare system and improving ease of use and reliability. Sending the monitoring firmware to a cloud platform facilitates centralized management and analysis of patient data. The cloud platform can process large amounts of data, conduct real-time risk assessments, and provide timely feedback to medical staff. The cloud platform provides remote access, allowing medical staff to access risk assessment results at any time, facilitating cross-regional and multi-institutional collaboration and decision-making, and improving the accessibility and responsiveness of medical services.Therefore, the present invention is an improvement on the traditional method of constructing a risk assessment model for myocardial infarction patients, which solves the problems of large errors in risk assessment of myocardial infarction patients and inaccurate identification of risk antecedent factors for myocardial infarction patients in the traditional method of constructing a risk assessment model for myocardial infarction patients. It reduces the error in risk assessment of myocardial infarction patients and improves the accuracy of identifying risk antecedent factors for myocardial infarction patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 A schematic diagram of the steps for constructing a risk assessment model for patients with myocardial infarction;

[0071] Figure 2 for Figure 1 The detailed implementation steps of step S2 are shown in FIG.

[0072] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0073] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0074] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0075] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0076] To achieve this, please refer to Figures 1 to 2 A method for constructing a risk assessment model for myocardial infarction patients comprises the following steps:

[0077] Step S1: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system to obtain vital sign record data and coronary angiography images of myocardial infarction patients, respectively; constructing an arterial angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients to obtain an arterial angiography multi-view 3D model;

[0078] Step S2: performing ECG signal-vasomotion mapping processing on the multi-view 3D arterial angiography model to obtain ECG signal-vasomotion mapping data; performing vascular occlusion increment calculation on the ECG signal-vasomotion mapping data to obtain vascular occlusion increment data; performing ECG signal pre-gradient feedback identification on the vascular occlusion increment data based on the ECG signal-vasomotion mapping data to obtain ECG pre-gradient feedback data;

[0079] Step S3: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; using a random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the blocked feedback ECG signal phase offset data to obtain a risk assessment model for myocardial infarction patients;

[0080] Step S4: Design an automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the risk assessment model monitoring firmware for myocardial infarction patients, and send the risk assessment model monitoring firmware for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

[0081] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a method for constructing a risk assessment model for myocardial infarction patients according to the present invention. In this example, the method for constructing a risk assessment model for myocardial infarction patients includes the following steps:

[0082] Step S1: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system to obtain vital sign record data and coronary angiography images of myocardial infarction patients, respectively; constructing an arterial angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients to obtain an arterial angiography multi-view 3D model;

[0083] In this embodiment of the present invention, a hospital's internal medical information system utilizes an integrated medical data acquisition module to collect various vital signs of patients with myocardial infarction, including blood pressure, heart rate, respiratory rate, and blood oxygen saturation. This data is acquired in real time by connected multi-functional monitoring devices (such as Holter monitors and arterial blood pressure monitors) and synchronized to a central database to generate a record of the vital signs of patients with myocardial infarction. Subsequently, a coronary angiography device performs angiography scan, and digital subtraction technology is used to obtain images of the coronary arteries from different angles. The acquired image data undergoes preprocessing, and a segmentation algorithm is used to extract the vascular structure. The coronary artery images from different angles are then matched and registered to construct a three-dimensional coronary artery model from multiple perspectives. During this process, an algorithm based on geometric reconstruction is used to refine the vascular structure, ensuring the authenticity and accuracy of the arterial model, ultimately generating a multi-view 3D arterial angiography model.

[0084] Step S2: performing ECG signal-vasomotion mapping processing on the multi-view 3D arterial angiography model to obtain ECG signal-vasomotion mapping data; performing vascular occlusion increment calculation on the ECG signal-vasomotion mapping data to obtain vascular occlusion increment data; performing ECG signal pre-gradient feedback identification on the vascular occlusion increment data based on the ECG signal-vasomotion mapping data to obtain ECG pre-gradient feedback data;

[0085] In this embodiment of the present invention, based on a multi-view 3D arterial angiography model and combined with ECG signal data, a specific ECG signal-vasomotion mapping algorithm first analyzes ECG signal waveform changes and spectral information to identify signal features related to hemodynamics in various cardiac regions. Using these features, the ECG signal is mapped to vasomotion, establishing a dynamic correlation between the ECG signal and vasomotion, and generating ECG signal-vasomotion mapping data. During this mapping process, computational fluid dynamics (CFD) technology is used to simulate blood flow within the arterial model, simulating the velocity and pressure changes of blood flowing through the coronary arteries, thereby enabling simultaneous correlation analysis of the ECG signal and vasomotion. Calculus analysis is applied to the resulting ECG signal-vasomotion mapping data to calculate incremental values ​​of localized vascular obstruction, generating incremental vascular obstruction data. The specific method employed in this process includes continuous monitoring of the dynamic changes in the coronary artery lumen diameter and, using a high-precision measurement system, real-time comparative analysis of blood flow velocities before and after the stenotic segment to calculate the changing trend and cumulative amount of vascular obstruction. Finally, based on the ECG signal-vasomotion mapping data, the incremental data for vascular obstruction is further processed using ECG signal pre-gradient feedback to identify obstruction. This process utilizes signal processing techniques to predict subtle changes in the ECG signal in advance and identifies and analyzes the signal pre-gradient through a multi-channel feedback system. This process involves extracting specific ECG signal bands associated with obstruction and performing time series analysis on these bands to identify potential precursors to cardiac obstruction, generating the final ECG pre-gradient feedback data for obstruction.

[0086] Step S3: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; using a random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the blocked feedback ECG signal phase offset data to obtain a risk assessment model for myocardial infarction patients;

[0087] In an embodiment of the present invention, phase analysis techniques are used to analyze the phase offset of the generated blocked ECG pre-gradient feedback data. The specific implementation steps include performing a Fourier transform on the ECG signal to convert the time-domain signal into a frequency-domain signal, thereby extracting the phase information of different frequency components in the ECG signal. Based on this, a Hilbert transform is applied to track the phase changes of the frequency components over time, identifying phase offsets associated with coronary artery obstruction. During the analysis process, a signal denoising algorithm is used to eliminate external interference and ensure the accuracy of the phase offset data. Finally, based on the offset results of each phase component, phase offset data of the blocked feedback ECG signal is generated. After obtaining the phase offset data, a random forest algorithm is used to construct a risk assessment model for patients with myocardial infarction. The construction process first performs feature selection on the phase offset data, using statistical methods to screen the phase features most closely associated with myocardial infarction risk. Subsequently, multiple decision trees are constructed, and the prediction results of these multiple decision trees are integrated using the random forest algorithm to achieve a comprehensive assessment of the risk of patients with myocardial infarction. In order to enhance the stability and generalization ability of the model, cross-validation was used to train and test the model to ensure that the model could accurately assess the risk levels of different patients and ultimately generate a risk assessment model for myocardial infarction patients.

[0088] Step S4: Design an automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the risk assessment model monitoring firmware for myocardial infarction patients, and send the risk assessment model monitoring firmware for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

[0089] In an embodiment of the present invention, after constructing a risk assessment model for myocardial infarction patients, automated monitoring firmware specifically designed for this model is designed. This firmware utilizes an embedded system design, integrating the model's operational logic and monitoring functions into hardware to enable real-time data monitoring and model inference. The firmware design process includes optimizing the sensor interface to ensure that ECG signals can be collected and transmitted to the firmware's computational unit in real time. Furthermore, the firmware integrates a signal processing module to compare the real-time ECG signals with the risk assessment model for myocardial infarction patients to determine changes in the patient's risk of myocardial infarction. After the firmware design is complete, it is debugged and optimized using compilation tools to ensure stable operation in different hardware environments, ultimately generating firmware for monitoring the risk assessment model for myocardial infarction patients. After the firmware design is complete, it is uploaded to a medical cloud platform. During the upload process, the firmware's security and stability are ensured through encryption technology to prevent data leakage or unauthorized access. On the cloud platform, the firmware monitors the patient's ECG data in real time and performs inference calculations based on the risk assessment model for myocardial infarction patients to assess their risk. This process ensures automated and continuous real-time monitoring and risk assessment of myocardial infarction patients, providing timely support for medical decision-making.

[0090] Preferably, step S1 includes the following steps:

[0091] Step S11: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system, and obtaining basic data of myocardial infarction patients including vital sign record data and coronary angiography images of myocardial infarction patients;

[0092] Step S12: performing image noise optimization on the coronary angiography image of the myocardial infarction patient to obtain a coronary artery noise optimized image;

[0093] Step S13: performing multi-view reconstruction analysis on the coronary angiography image of the myocardial infarction patient based on the coronary artery noise optimization image to obtain coronary angiography multi-view reconstruction data;

[0094] Step S14: constructing an artery angiography multi-view 3D model based on the coronary angiography multi-view reconstruction data to obtain an artery angiography multi-view 3D model.

[0095] In this embodiment of the present invention, multiple data acquisition modules are sequentially accessed through the hospital's medical information system and connected to MI patient monitoring equipment to collect vital sign data in real time. Vital sign data, including heart rate, blood pressure, blood oxygen saturation, and respiratory rate, are recorded in real time by a multi-lead electrocardiograph, a blood pressure monitor, and an oxygen saturation monitor to generate a MI patient vital sign record. Simultaneously, the patient enters the catheterization laboratory, where coronary artery angiography is performed using digital subtraction angiography equipment to obtain high-resolution coronary artery imaging data. After acquisition, all image data is automatically stored in an image database and synchronized with the vital sign data, forming a baseline MI patient data set consisting of MI patient vital sign records and MI coronary angiography images. The collected coronary angiography image data undergoes noise optimization. First, an image processing algorithm is used to filter out high-frequency noise in the image. A noise reduction algorithm based on adaptive filtering is employed to preserve image edge details and remove clutter caused by X-ray scattering. During the optimization process, a post-noise enhancement approach is employed to readjust the grayscale distribution of the image, enhancing the clarity of coronary artery details. The final processing result is a noise-optimized coronary artery image. This image eliminates noise while ensuring image accuracy and clarity, providing a high-quality data foundation for subsequent analysis. Multi-view reconstruction analysis is performed on the noise-optimized coronary artery image. First, image registration techniques are used to align coronary angiography images from different angles to ensure consistency across different viewpoints. Next, image segmentation techniques are used to accurately segment the coronary artery structure in each frame, generating contours of the vascular structure. By integrating the segmentation results from multiple viewpoints, a volume rendering algorithm and multi-angle projection algorithms are used to perform 3D reconstruction analysis of the coronary arteries, generating multi-view reconstruction data for the coronary angiography. This step ensures the integrity and accuracy of the coronary arteries from different angles, providing a reliable data foundation for 3D model construction. Based on the multi-view reconstruction data, a multi-view 3D model of the coronary artery is constructed. First, based on the reconstructed data, volume reconstruction techniques are used to reconstruct the coronary artery lumen in 3D, establishing the complete arterial geometry. To ensure model accuracy, finite element analysis was used to simulate the hemodynamics of the coronary artery model and calibrate the model's geometric deviations. Finally, using a segmented reconstruction method, the different segmented 3D models were spliced ​​and fused to create a complete multi-view 3D arterial angiographic model. This model accurately displays the three-dimensional structure of the coronary arteries, providing support for subsequent vascular analysis and diagnosis.

[0096] Preferably, step S2 includes the following steps:

[0097] Step S21: extracting electrocardiogram records from the myocardial infarction patient's physical sign record data to obtain an electrocardiogram of the myocardial infarction patient;

[0098] Step S22: performing electrocardiogram-vascular motion mapping processing on the multi-view 3D model of arteriography according to the electrocardiogram of the myocardial infarction patient to obtain electrocardiogram-vascular motion mapping data;

[0099] Step S23: using a preset myocardial abnormal behavior recognition model to perform myocardial abnormal behavior recognition on the ECG signal-vasomotion mapping data, to obtain myocardial abnormal behavior data of the myocardial infarction patient;

[0100] Step S24: calculating the vascular occlusion increment on the ECG signal-vascular motion mapping data according to the myocardial abnormal behavior data of the myocardial infarction patient to obtain vascular occlusion increment data;

[0101] Step S25: performing blockage ECG signal pre-gradient feedback identification on the vascular blockage increment data according to the ECG signal-vascular motion mapping data to obtain blockage ECG pre-gradient feedback data.

[0102] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0103] Step S21: extracting electrocardiogram records from the myocardial infarction patient's physical sign record data to obtain an electrocardiogram of the myocardial infarction patient;

[0104] In an embodiment of the present invention, electrocardiogram records are extracted from the vital sign record data of myocardial infarction patients. First, the data is preprocessed using the original electrocardiogram signal obtained by a multi-lead electrocardiograph. The preprocessing step includes removing baseline drift based on high-pass filtering technology, and using low-pass filtering to eliminate myoelectric noise interference. Next, the R peak detection algorithm is used to locate the key heartbeat events in the electrocardiogram, and the electrocardiogram signal is segmented according to the RR interval for subsequent analysis. The preprocessed signal is stored in a standardized electrocardiogram data format, and finally the electrocardiogram data of the myocardial infarction patient is generated. The electrocardiogram record will be used for subsequent correlation analysis of the electrocardiogram signal and coronary artery movement.

[0105] Step S22: performing electrocardiogram-vascular motion mapping processing on the multi-view 3D model of arteriography according to the electrocardiogram of the myocardial infarction patient to obtain electrocardiogram-vascular motion mapping data;

[0106] In an embodiment of the present invention, an electrocardiogram-vascular motion mapping process is performed on the constructed multi-view three-dimensional model of arterial angiography based on the electrocardiogram of a myocardial infarction patient. First, the cardiac periodic signal in the electrocardiogram is matched with the corresponding vascular motion in the coronary artery model, and the key features of the electrocardiogram, such as the amplitude and duration of the QRS complex and the T wave, are extracted by time domain analysis technology. Then, a frequency domain analysis technology based on Fourier transform is used to calculate the spectral distribution of the electrocardiogram signal, and this information is mapped to the motion characteristics of the coronary artery to form the spatiotemporal correlation data of the electrocardiogram-vascular motion. This process needs to comprehensively consider the synchronization between cardiac contraction and vascular wall movement to ensure the accuracy of the mapping. The final electrocardiogram-vascular motion mapping data contains the correlation information between electrocardiogram activity and the dynamic changes of the coronary artery.

[0107] Step S23: using a preset myocardial abnormal behavior recognition model to perform myocardial abnormal behavior recognition on the ECG signal-vasomotion mapping data, to obtain myocardial abnormal behavior data of the myocardial infarction patient;

[0108] In an embodiment of the present invention, a preset myocardial abnormal behavior recognition model is used to analyze the ECG signal-vascular motion mapping data to identify abnormal myocardial behavior in patients with myocardial infarction. The model uses a rule-based myocardial abnormal behavior recognition algorithm, which mainly identifies the behavioral patterns of lesions such as myocardial ischemia and myocardial fibrosis. First, through time series analysis of the mapping data, characteristic signals of abnormal myocardial behavior, such as abnormal contraction and relaxation patterns, are extracted. Then, using a feature matching algorithm, these features are compared with the abnormal behavior templates preset in the model to identify potential abnormal myocardial behavior. The recognition results include the specific type of abnormal behavior, the time period of occurrence, and the scope of influence, and ultimately generate abnormal myocardial behavior data for patients with myocardial infarction for further risk assessment.

[0109] Step S24: calculating the vascular occlusion increment on the ECG signal-vascular motion mapping data according to the myocardial abnormal behavior data of the myocardial infarction patient to obtain vascular occlusion increment data;

[0110] In an embodiment of the present invention, the ECG signal-vascular motion mapping data is used to calculate the incremental vascular obstruction based on the abnormal myocardial behavior data of patients with myocardial infarction. First, the specific vascular area involved in the abnormal myocardial behavior data is extracted, and the degree of vascular obstruction in this area is evaluated by calculating hemodynamic parameters such as vascular resistance, flow velocity and pressure difference. Then, based on the dynamic motion information in the mapping data, a differential calculation method is used to determine the temporal trend of the degree of obstruction, that is, the incremental change of vascular obstruction. This process requires nonlinear regression analysis to quantify the impact of ECG signal changes on the degree of vascular obstruction and generate accurate incremental vascular obstruction data. This data is used to predict the progression of vascular obstruction and guide subsequent treatment decisions.

[0111] Step S25: performing blockage ECG signal pre-gradient feedback identification on the vascular blockage increment data according to the ECG signal-vascular motion mapping data to obtain blockage ECG pre-gradient feedback data.

[0112] In an embodiment of the present invention, based on the ECG signal-vascular motion mapping data, the ECG signal pre-gradient feedback identification of vascular obstruction incremental data is performed. First, the vascular obstruction degree change curve in the blockage incremental data is used in combination with the dynamic changes of the ECG signal to calculate the gradient change characteristics of the ECG signal, and identify the ECG signal pre-gradient abnormality caused by the blockage. By constructing a feedback model based on the ECG signal gradient change, the response pattern of the ECG signal under different blockage levels is analyzed to determine the critical point of the gradient change. Then, using the reverse deduction algorithm, the abnormal gradient is fed back to the early ECG signal to identify potential early signs of blockage and generate ECG pre-gradient feedback data of blockage. This data is used to evaluate the impact of the degree of blockage on cardiac function and provide timely early warning information.

[0113] Preferably, step S24 includes the following steps:

[0114] Step S241: calculating abnormal cardiac diastolic frequency based on abnormal myocardial behavior data of myocardial infarction patients to obtain abnormal cardiac diastolic frequency data;

[0115] Step S242: performing blood transport attenuation dynamics simulation on the ECG signal-vascular motion mapping data according to the abnormal diastolic frequency data to obtain blood transport attenuation dynamics simulation data;

[0116] Step S243: performing pressure increment azimuth decomposition calculation on the blood transport attenuation dynamics simulation data to obtain blood pressure increment azimuth decomposition data;

[0117] Step S244: Calculate the vascular occlusion increment based on the blood pressure increment azimuth decomposition data to obtain vascular occlusion increment data.

[0118] In an embodiment of the present invention, when calculating abnormal cardiac diastolic frequency based on abnormal myocardial behavior data from patients with myocardial infarction, the diastolic phase of the abnormal myocardial region is first extracted from the ECG signal. In this specific operation, the start and end moments of myocardial diastole are located using time-domain analysis methods. Then, a fast Fourier transform (FFT) is applied to convert the time signal into a frequency-domain signal, and the fluctuation of the abnormal frequency is analyzed. By calculating the amplitude of diastolic frequency fluctuation within each cardiac cycle and combining it with the time points of abnormal events in the abnormal myocardial behavior data, the overall abnormal cardiac diastolic frequency data is calculated. This data reflects the abnormal state of myocardial diastole and is used in subsequent dynamic simulation of blood transport attenuation. Based on the abnormal cardiac diastolic frequency data, a dynamic simulation of blood transport attenuation is performed on the ECG signal-vascular motion mapping data. First, using hemodynamic equations and combining the abnormal values ​​in the diastolic frequency data, a resistance and attenuation model for blood transport is constructed. In the specific implementation, the interaction between changes in vessel wall elasticity and blood viscosity is calculated to determine the blood transport efficiency in the coronary arteries. Finite element analysis is used to simulate the trajectory and velocity of blood near the obstruction area, quantifying the energy loss during this process. This generates dynamic simulation data for blood transport attenuation. This data reflects the attenuation of blood transport capacity under obstruction. To perform azimuthal decomposition of the pressure increments on this data, the researchers first extract blood pressure variations at different locations within the coronary artery based on the attenuation dynamics data. A pressure increment calculation matrix is ​​then established to decompose the blood pressure in each direction. Tensor decomposition techniques are then used to analyze the pressure differences and trends across the vessel. Specifically, the blood pressure data is divided into three directions: radial, axial, and tangential. Pressure increments are calculated for each direction, and a pressure increment distribution map is generated using analytical geometry. The azimuthal decomposition results reflect the pressure distribution at different vascular locations, generating azimuthal decomposition data for subsequent vascular obstruction increment calculations. To calculate vascular obstruction increments based on the azimuthal decomposition data, a model for calculating the degree of vascular obstruction is first constructed using the pressure distribution data. By combining the pressure increment data with the elastic modulus of the vessel wall, the Laplace equation is used to calculate the deformation of the vessel wall under varying pressures. Subsequently, through integration, the degree of blockage at different locations in the blood vessels is quantified, generating a dynamic curve of blockage. Finally, based on this data, incremental vascular blockage data is generated to accurately describe the progression of blockage within the coronary arteries of patients with myocardial infarction. This data provides a critical reference for subsequent clinical interventions and treatment plans.

[0119] Preferably, performing pressure increment azimuthal decomposition calculation on the blood transport attenuation dynamics simulation data comprises the following steps:

[0120] Performing transport rate attenuation evaluation on the blood transport attenuation dynamics simulation data to obtain blood transport rate attenuation data;

[0121] Performing blood fluidity pressure intensity simulation based on blood transport rate attenuation data to obtain blood fluidity pressure intensity data;

[0122] Performing omnidirectional stress tensor calculation on the blood fluidity pressure intensity data to obtain omnidirectional stress tensor data;

[0123] Based on the omnidirectional stress tensor data, stress spatial distribution fluctuation analysis is performed to obtain stress spatial distribution fluctuation data;

[0124] The spatial potential energy relationship of stress spatial distribution fluctuation data is quantified according to the Bernoulli equation to obtain stress spatial potential energy relationship data;

[0125] The pressure increment azimuth decomposition calculation is performed based on the stress space potential energy relationship data and the stress space distribution fluctuation data to obtain the blood pressure increment azimuth decomposition data.

[0126] In an embodiment of the present invention, when evaluating transport velocity attenuation using dynamic simulation data for blood transport attenuation, the blood velocity attenuation trend in the obstructed area is first calculated using the blood velocity curve in the coronary arteries in conjunction with a pulsatile flow model. Specifically, by analyzing the blood flow velocity variations in each branch of the coronary artery, fluid dynamics equations are used to quantitatively analyze the blood velocity attenuation, yielding velocity attenuation values ​​at different locations. To ensure accurate evaluation, a piecewise integration method is used to calculate the average blood velocity variation over different time periods, generating complete blood transport velocity attenuation data. This data provides the basis for subsequent simulation of blood flow pressure intensity. The blood flow pressure intensity simulation is performed based on the blood transport velocity attenuation data. Specifically, the fluid dynamics pressure formula is used to link blood flow velocity with the pressure intensity experienced by the vessel wall. First, a stepwise linear regression analysis is performed on the flow velocity attenuation data to extract the degree to which blood flow velocity changes affect pressure. Then, computational fluid dynamics (CFD) methods are used to simulate the pressure field of blood flow in the coronary arteries, generating pressure distribution maps at different locations within the vessel. During the pressure intensity simulation, the effects of blood viscosity and vascular elasticity are considered to generate blood flow pressure intensity data. This data reflects the pressure distribution of blood under obstruction and lays the foundation for stress tensor calculation. To calculate the omnidirectional stress tensor based on blood flow pressure intensity data, the pressure components in different directions within the vessel are first decomposed based on the three-dimensional stress tensor formula. Specifically, a stress field model is established, and the finite element method is used to calculate the pressure at each node within the vessel point by point, extracting the principal stress components in three-dimensional space. During this process, Lagrange interpolation is used to smooth the data to ensure the continuity and stability of the calculation results. By accumulating each stress tensor within the coronary artery item by item, omnidirectional stress tensor data is obtained. This data describes the stress distribution in different directions within the vessel and provides a basis for subsequent spatial distribution fluctuation analysis. To perform stress spatial distribution fluctuation analysis based on the omnidirectional stress tensor data, the stress fluctuation amplitude in each direction is first calculated to construct a stress fluctuation matrix. Fast Fourier transform (FFT) is used to convert the stress signal in the time domain to the frequency domain, and the fluctuation amplitude in the frequency domain signal is analyzed. Then, combining the stress data from different regions within the coronary artery, the spatial stress fluctuation gradient is calculated using the difference method. The results of the stress spatial distribution fluctuation analysis generate stress spatial distribution fluctuation data, describing the degree of stress fluctuation in different regions within the coronary arteries. This data provides an analytical basis for quantifying spatial potential energy relationships. To quantify the spatial potential energy relationship of stress spatial distribution fluctuation data using the Bernoulli equation, the potential energy formula is first used to relate stress fluctuations to potential energy changes. Specifically, this involves utilizing the energy conservation principle of the Bernoulli equation to quantify the relationship between stress fluctuation data and blood kinetic energy and potential energy.By calculating the potential energy differences of blood at different spatial positions and combining the mutual conversion rules of kinetic energy and potential energy, stress space potential energy relationship data is generated. This data reveals the energy state of different parts in the coronary artery and provides a quantitative basis for the subsequent pressure increment azimuthal decomposition calculation. When performing pressure increment azimuthal decomposition calculation based on stress space potential energy relationship data and stress space distribution fluctuation data, the pressure increments in each direction are first matrix decomposed. The specific operation includes using the eigenvalue decomposition method in linear algebra to decompose the complex stress space matrix into incremental pressure fields in different directions. By gradually decomposing the pressure fields in each direction of the coronary artery and combining the potential energy relationship data, the pressure increment in each direction is weighted. The blood pressure increment azimuthal decomposition data finally generated clearly shows the pressure changes in different directions of the coronary artery, providing important basic data for further risk assessment.

[0127] Preferably, step S25 includes the following steps:

[0128] Step S251: performing time domain feature analysis on the ECG signal-vascular motion mapping data to obtain ECG signal-vascular motion time domain data;

[0129] Step S252: performing multi-factor correlation analysis on the vascular occlusion increment data based on the ECG signal-vascular motion time domain data to obtain multi-factor correlation data of occlusion;

[0130] Step S253: performing multi-factor cluster analysis on the blocking multi-factor correlation data to obtain blocking multi-factor correlation cluster data;

[0131] Step S254: performing dynamic blockage response simulation on the vascular blockage increment data based on the blockage multi-factor association clustering data to obtain multi-factor dynamic blockage response data;

[0132] Step S255: performing signal response pre-gradient trend identification on the ECG signal-vasomotion mapping data according to the multi-factor dynamic occlusion response data to obtain signal response pre-gradient trend data;

[0133] Step S256: performing blocked ECG signal pre-gradient feedback identification according to the signal response pre-gradient trend data and the multi-factor dynamic blocking response data to obtain blocked ECG signal pre-gradient feedback data.

[0134] In this embodiment of the present invention, when performing time-domain feature analysis on ECG signal-vasomotion mapping data, high-resolution time-domain signal processing techniques are first used to refine the specific manifestations of ECG signals in vasomotion. This process involves segmenting the signal using short-time Fourier transform (STFT) to analyze the temporal correlation between ECG signals and vasomotion in different time periods. Feature extraction is then performed for each time-domain segment, primarily for signal amplitude, frequency, phase, and other variation parameters. By comparing the volatility and regularity of these parameters, ECG signal-vasomotion time-domain data is generated. This data serves as input for subsequent multi-factor association analysis of obstruction, providing temporal variation characteristics between ECG signals and vasomotion. When performing multi-factor association analysis of vascular obstruction increment data based on the ECG signal-vasomotion time-domain data, linear regression analysis combined with principal component analysis (PCA) is used to explore correlations between ECG signal time-domain features and multiple factors contributing to vascular obstruction increment. Specifically, a multivariate linear regression model is constructed to quantify the correlation between characteristic points in the ECG signal and the degree of vascular obstruction. PCA technology was also used to reduce dimensionality and extract the most significant influencing factors, ultimately generating multi-factor association data for blockage. This process, by refining the correlations between multiple factors, revealed the complex interactions between ECG signals and changes in vascular obstruction, laying the foundation for subsequent cluster analysis. The multi-factor cluster analysis of the blockage multi-factor association data began with a K-means clustering algorithm to classify the multi-factor data and identify characteristic patterns under different blockage states. This process involved normalizing the multi-factor association data to ensure uniform data scale. The clustering algorithm then divided the data into multiple clusters, each representing a specific blockage state. Cluster centers were then calculated using Euclidean distance to ensure similarity within each cluster while maximizing inter-cluster variance. By gradually optimizing the clustering results, multi-factor association cluster data were generated, which clearly reflected the distribution of multi-factor characteristics under different blockage states. Dynamic blockage response simulations of incremental vascular blockage data based on the multi-factor association cluster data were performed using dynamic system modeling methods to simulate the blockage process. The specific operation involves using dynamic simulation tools (such as MATLAB's Simulink module) to map clustered data onto a time series of vascular obstruction increments, simulating the progression of vascular obstruction under different external factors. By simulating and analyzing the evolution of obstruction increments over time, multi-factor dynamic obstruction response data is generated. This data reveals the dynamic response process under different obstruction scenarios and provides a basis for subsequent signal response pre-gradient trend identification. When using multi-factor dynamic obstruction response data to identify signal response pre-gradient trends in ECG signal-vascular motion mapping data, the first step is to extract the gradient trend by calculating the obstruction response data's representation in the ECG signal.The specific operation involves using a gradient descent algorithm to capture and analyze subtle fluctuations between ECG signals and vascular motion, and using curve fitting to predict future signal trends. By gradually adjusting gradient parameters to optimize trend prediction accuracy, the system ultimately generates signal response pre-gradient trend data. This data reflects subtle changes in the ECG signal before occlusion, providing a theoretical basis for predicting the onset of occlusion. When performing blockage ECG signal pre-gradient feedback identification based on this data and multi-factor dynamic blockage response data, a feedback loop is constructed to dynamically track the relationship between ECG signal gradient changes and blockage responses. The system employs feedback control theory to design a closed-loop feedback system that detects pre-gradients in the ECG signal in real time and feeds the detection results into the blockage response model for adjustment. By gradually optimizing feedback parameters to ensure coordination between the ECG signal and vascular blockage response, the system ultimately generates blockage ECG pre-gradient feedback data. This data provides a key predictive basis for risk assessment models for myocardial infarction patients, enabling effective early warning of blockage risk through accurate pre-gradient identification.

[0135] Preferably, step S3 includes the following steps:

[0136] Step S31: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data;

[0137] Step S32: performing spatial interpolation processing on the blocked feedback ECG signal phase offset data to obtain signal offset spatial interpolation data;

[0138] Step S33: performing signal compensation on the blocked feedback ECG signal phase offset data according to the signal offset spatial interpolation data to obtain blocked feedback signal phase offset compensation data;

[0139] Step S34: Using the random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the phase offset compensation data of the blocking feedback signal, a risk assessment model for myocardial infarction patients is obtained.

[0140] In an embodiment of the present invention, when analyzing ECG signal phase offset for blocked ECG pre-gradient feedback data, the ECG signal is first converted from the time domain to the frequency domain using a Fourier transform to extract the signal's phase information. Next, a phase analysis technique (such as the Hilbert transform) is used to perform a detailed analysis of the signal's phase offset to identify phase changes during the blocking process. This analysis focuses on identifying the amplitude and trend of the signal phase offset, particularly in specific ECG waveform regions. By calculating the amplitude, frequency, and duration of the phase change, blocked ECG feedback signal phase offset data is obtained, providing fundamental information for subsequent signal compensation and risk assessment. When performing spatial interpolation processing on the blocked ECG feedback signal phase offset data, the data points are first spatially gridded to form a regular grid structure. Interpolation algorithms (such as bilinear interpolation or cubic spline interpolation) are then used to estimate and fill the gaps between these data points. This process involves selecting an appropriate interpolation method, determining the neighborhood of the data points, and performing spatial interpolation to obtain estimated values ​​at the gaps between the data points. This step aims to generate a smooth signal offset distribution that more accurately reflects signal spatial variations. The resulting spatial interpolation of signal offsets provides a smooth and uniform spatial information foundation for signal compensation. When compensating the phase offset data of the blocked feedback ECG signal based on the spatial interpolation of signal offsets, the interpolated data is first compared with the original ECG signal phase offset data to determine the direction and magnitude of the offset compensation. Compensation factors are then calculated and applied to the original data to adjust the phase offset in the signal. This involves constructing a compensation model and applying the interpolation results as correction data to the ECG signal to correct the offset. This process optimizes the compensation parameters to ensure that the compensated signal is more consistent with actual ECG activity, resulting in the phase offset-compensated data for the blocked feedback signal. This data provides corrected signal information for subsequent risk assessment. When constructing a risk assessment model for myocardial infarction patients using the random forest algorithm using the phase offset-compensated data for blocked feedback signals, the compensated data is first used as input features, and a dataset with known MI risk levels is prepared as training data. The random forest algorithm trains this data by constructing multiple decision trees, each using a different subset of random features to improve model accuracy and robustness. During the training process, the predictions of each decision tree are voted on to arrive at the final risk assessment result. Once the model is trained, it is applied to new ECG signal data to predict the risk level of myocardial infarction, ultimately resulting in a risk assessment model for myocardial infarction patients. This model accurately identifies the risk of myocardial infarction and supports clinical decision-making and intervention.

[0141] Preferably, step S32 includes the following steps:

[0142] Step S321: calculating the phase offset of the blocked feedback ECG signal phase offset data to obtain the blocked ECG signal phase offset data;

[0143] Step S322: performing nonlinear incremental calculation on the blocked feedback ECG signal phase offset data according to the blocked ECG signal phase offset data to obtain phase offset nonlinear incremental data;

[0144] Step S323: performing spatial increment trend analysis on the phase offset data of the blocked feedback ECG signal according to the phase offset nonlinear increment data to obtain phase offset spatial increment trend data;

[0145] Step S324: performing spatial interpolation processing on the phase offset data of the blocked feedback ECG signal based on the phase offset spatial increment trend data to obtain signal offset spatial interpolation data.

[0146] In an embodiment of the present invention, when calculating the phase offset of blocked feedback ECG signal phase offset data, the phase value at each time point is first extracted from the ECG signal phase information. Next, the phase offset at each time point is calculated by calculating the phase difference between adjacent time points. The specific operation includes converting the signal phase data into time series data, then performing a subtraction operation on each pair of adjacent points in the time series data to calculate the phase offset. During the calculation process, particular attention must be paid to phase periodicity to prevent calculation errors from affecting the final result. The resulting blocked ECG signal phase offset data provides quantitative information on the phase offset for further analysis. When performing nonlinear incremental calculation of the blocked feedback ECG signal phase offset data based on the blocked ECG signal phase offset data, the changing trend of the phase offset data is first determined. A nonlinear incremental calculation method (such as polynomial fitting or curve fitting) is then used to fit the phase offset time series data to extract nonlinear variation characteristics. The specific steps include fitting the phase offset data to a nonlinear function using a nonlinear regression model and calculating the model increment. This operation aims to identify and quantify the nonlinear variation trend in the signal and obtain nonlinear incremental phase offset data. These data help more accurately describe the complex signal variation patterns. When performing spatial incremental trend analysis on the phase offset data of blocked feedback ECG signals based on nonlinear phase offset incremental data, the spatial distribution of the nonlinear phase offset incremental data is first analyzed. By calculating the incremental value at each spatial location, spatial incremental trend modeling is then constructed using spatial interpolation techniques (such as Kriging interpolation). This involves associating the spatial locations of data points with the corresponding nonlinear incremental data, constructing a spatial incremental trend model, and calculating the incremental trend at each location. The resulting spatial incremental trend data for phase offset provides the foundation for further spatial interpolation processing, helping to understand the spatial variation patterns of the signal. When performing spatial interpolation on the phase offset data of blocked feedback ECG signals based on the spatial incremental trend data, an appropriate spatial interpolation algorithm (such as cubic spline interpolation or bilinear interpolation) is first selected to interpolate between data points. Using the spatial incremental trend data as the interpolation basis, an interpolation function is constructed to calculate estimated values ​​between data points. The specific steps include interpolating the spatial incremental trend data on a spatial grid, obtaining an estimated value at each grid point, and integrating this estimated value into the ECG phase offset data. The resulting signal offset spatial interpolation data provides smooth phase information, which is helpful for further analysis and processing.

[0147] Preferably, step S34 includes the following steps:

[0148] Step S341: dividing the blocking feedback signal phase offset compensation data into a data set to obtain a blocking feedback signal phase offset compensation test set and a blocking feedback signal phase offset compensation training set;

[0149] Step S342: performing feedback signal similarity feature selection on the blocking feedback signal phase offset compensation training set to obtain feedback signal similarity feature data;

[0150] Step S343: using a random forest algorithm and performing similar structure incremental learning on the blocking feedback signal phase offset compensation training set based on the similar feature data of the feedback signal to obtain similar structure incremental data;

[0151] Step S344: constructing an initial risk assessment model for myocardial infarction patients based on the similar structure incremental data to obtain an initial risk assessment model for myocardial infarction patients;

[0152] Step S345: performing a model test on the initial myocardial infarction patient risk assessment model based on the blocking feedback signal phase offset compensation test set to obtain a myocardial infarction patient risk assessment model.

[0153] In an embodiment of the present invention, when partitioning the data set for the phase offset compensation data of the blocking feedback signal, the data is first divided proportionally into a training set and a test set. Specifically, this operation includes randomly allocating the phase offset compensation data according to a predetermined ratio (e.g., 80% training set, 20% test set) to ensure that each subset represents the characteristic distribution of the overall data. During the partitioning process, data leakage is avoided and overlap between the training and test sets is ensured. The training set is used for model training, while the test set is used for model performance verification. This partitioning operation ensures the model's ability to generalize to new data, resulting in a testing set and a training set for the phase offset compensation of the blocking feedback signal. When selecting similar features of the feedback signal from the phase offset compensation training set, feature data is first extracted from the signal and feature selection is performed. Specifically, the steps include calculating the eigenvalue of each signal data point and applying a feature selection algorithm (such as information gain or chi-square test) to select features with strong correlation with the ECG signal. The most representative feature data is selected by comparing the similarity metrics between each feature and the ECG signal. Ultimately, similar feature data of the feedback signal is obtained, which is used to train the model and improve its prediction accuracy. When using the random forest algorithm to perform incremental similarity learning on the training set for phase offset compensation of blocked feedback signals based on similarity feature data of the feedback signal, a random forest model is first constructed. This involves quantizing the features of each sample in the training set and inputting the similarity feature data of the feedback signal into the random forest model. The random forest model then builds multiple decision trees and uses a voting mechanism to determine the final incremental learning result. This process includes feature importance assessment and incremental learning. The resulting incremental similarity data is used for further model construction, helping to improve the model's ability to predict myocardial infarction risk. When constructing an initial risk assessment model for myocardial infarction patients based on the incremental similarity data, the incremental similarity data is first integrated into the risk assessment model. This involves using the aforementioned incremental data to update model parameters and construct a preliminary risk assessment model. This model is then fitted to the samples in the training set to adjust the model's weights and biases. The construction process includes model training, validation, and optimization to ensure that the model accurately predicts the risk level of myocardial infarction patients. Ultimately, an initial risk assessment model for myocardial infarction patients is obtained, providing a foundation for subsequent testing. When testing the initial MI patient risk assessment model based on the phase offset compensation test set for blocked feedback signals, the test set data is first input into the established risk assessment model. This involves using the test set to make predictions for the model and evaluating its performance on real-world data. By calculating the difference between the predicted results and the true labels, the model's performance metrics, such as precision, recall, and F1 score, are evaluated. This process includes preprocessing the test data, generating the model's prediction output, and evaluating its performance.The obtained risk assessment model for myocardial infarction patients reflects the actual application effect of the model and ensures the effectiveness of the model in clinical settings.

[0154] Preferably, the present invention further provides a system for constructing a risk assessment model for myocardial infarction patients, which is used to execute the method for constructing a risk assessment model for myocardial infarction patients as described above. The system for constructing a risk assessment model for myocardial infarction patients comprises:

[0155] The angiography 3D model construction module is used to collect vital sign data and coronary angiography images of myocardial infarction patients through the medical information system, thereby obtaining vital sign records of myocardial infarction patients and coronary angiography images of myocardial infarction patients, respectively; and construct an angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients, thereby obtaining an angiography multi-view 3D model.

[0156] The ECG signal pre-gradient feedback recognition module is used to perform ECG signal-vascular motion mapping processing on the multi-view 3D model of arteriography to obtain ECG signal-vascular motion mapping data; perform vascular blockage increment calculation on the ECG signal-vascular motion mapping data to obtain vascular blockage increment data; and perform ECG signal pre-gradient feedback recognition on the vascular blockage increment data based on the ECG signal-vascular motion mapping data to obtain ECG pre-gradient feedback data;

[0157] The myocardial infarction patient risk assessment model construction module is used to perform ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; a myocardial infarction patient risk assessment model is constructed using the blocked feedback ECG signal phase offset data using a random forest algorithm to obtain a myocardial infarction patient risk assessment model;

[0158] The automated monitoring operation module is used to design automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the monitoring firmware for the risk assessment model for myocardial infarction patients, and send the monitoring firmware for the risk assessment model for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

[0159] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0160] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a risk assessment model for myocardial infarction patients, characterized in that: The following steps are involved: Step S1: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system to obtain vital sign record data and coronary angiography images of myocardial infarction patients, respectively; constructing an arterial angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients to obtain an arterial angiography multi-view 3D model; Step S2: performing ECG signal-vasomotion mapping processing on the multi-view 3D arterial angiography model to obtain ECG signal-vasomotion mapping data; performing vascular occlusion increment calculation on the ECG signal-vasomotion mapping data to obtain vascular occlusion increment data; performing occlusion ECG signal pre-gradient feedback identification on the vascular occlusion increment data based on the ECG signal-vasomotion mapping data to obtain occlusion ECG pre-gradient feedback data. Step S2 includes the following steps: Step S21: extracting electrocardiogram records from the myocardial infarction patient's physical sign record data to obtain an electrocardiogram of the myocardial infarction patient; Step S22: performing electrocardiogram-vascular motion mapping processing on the multi-view 3D model of arteriography according to the electrocardiogram of the myocardial infarction patient to obtain electrocardiogram-vascular motion mapping data; Step S23: using a preset myocardial abnormal behavior recognition model to perform myocardial abnormal behavior recognition on the ECG signal-vasomotion mapping data, to obtain myocardial abnormal behavior data of the myocardial infarction patient; Step S24: calculating the vascular occlusion increment on the ECG signal-vascular motion mapping data according to the myocardial abnormal behavior data of the myocardial infarction patient to obtain vascular occlusion increment data; Step S25: performing blockage ECG signal pre-gradient feedback identification on the vascular blockage increment data according to the ECG signal-vascular motion mapping data to obtain blockage ECG pre-gradient feedback data. Step S25 includes the following steps: Step S251: performing time domain feature analysis on the ECG signal-vascular motion mapping data to obtain ECG signal-vascular motion time domain data; Step S252: performing blockage multi-factor correlation analysis on the vascular blockage increment data based on the ECG signal-vascular motion time domain data to obtain blockage multi-factor correlation data; Step S253: performing multi-factor cluster analysis on the blocking multi-factor correlation data to obtain blocking multi-factor correlation cluster data; Step S254: performing dynamic blockage response simulation on the vascular blockage increment data based on the blockage multi-factor association clustering data to obtain multi-factor dynamic blockage response data; Step S255: performing signal response pre-gradient trend identification on the ECG signal-vasomotion mapping data according to the multi-factor dynamic occlusion response data to obtain signal response pre-gradient trend data; Step S256: performing blocked ECG signal pre-gradient feedback identification based on the signal response pre-gradient trend data and the multi-factor dynamic blocking response data to obtain blocked ECG signal pre-gradient feedback data; Step S3: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; using a random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the blocked feedback ECG signal phase offset data to obtain a risk assessment model for myocardial infarction patients. Step S3 includes the following steps: Step S31: performing an ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; Step S32: performing spatial interpolation processing on the blocked feedback ECG signal phase offset data to obtain signal offset spatial interpolation data; Step S33: performing signal compensation on the blocked feedback ECG signal phase offset data according to the signal offset spatial interpolation data to obtain blocked feedback signal phase offset compensation data; Step S34: using a random forest algorithm to construct a risk assessment model for myocardial infarction patients based on the phase offset compensation data of the blocking feedback signal, thereby obtaining a risk assessment model for myocardial infarction patients; Step S4: Design an automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the risk assessment model monitoring firmware for myocardial infarction patients, and send the risk assessment model monitoring firmware for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

2. The method for constructing a risk assessment model for myocardial infarction patients according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: collecting vital sign data and coronary angiography images of myocardial infarction patients through a medical information system, and obtaining basic data of myocardial infarction patients including vital sign record data and coronary angiography images of myocardial infarction patients; Step S12: performing image noise optimization on the coronary angiography image of the myocardial infarction patient to obtain a coronary artery noise optimized image; Step S13: performing multi-view reconstruction analysis on the coronary angiography image of the myocardial infarction patient based on the coronary artery noise point optimized image to obtain coronary angiography multi-view reconstruction data; Step S14: constructing an artery angiography multi-view 3D model based on the coronary angiography multi-view reconstruction data to obtain an artery angiography multi-view 3D model.

3. The method for constructing a risk assessment model for myocardial infarction patients according to claim 1, characterized in that: Step S24 includes the following steps: Step S241: calculating abnormal cardiac diastolic frequency based on abnormal myocardial behavior data of myocardial infarction patients to obtain abnormal cardiac diastolic frequency data; Step S242: performing blood transport attenuation dynamics simulation on the ECG signal-vascular motion mapping data according to the abnormal diastolic frequency data to obtain blood transport attenuation dynamics simulation data; Step S243: performing pressure increment azimuth decomposition calculation on the blood transport attenuation dynamics simulation data to obtain blood pressure increment azimuth decomposition data; Step S244: Calculate the vascular occlusion increment based on the blood pressure increment azimuth decomposition data to obtain vascular occlusion increment data.

4. The method for constructing a risk assessment model for myocardial infarction patients according to claim 3, characterized in that: The pressure increment azimuthal decomposition calculation of the blood transport attenuation dynamics simulation data includes the following steps: Performing transport rate attenuation evaluation on the blood transport attenuation dynamics simulation data to obtain blood transport rate attenuation data; Performing blood fluidity pressure intensity simulation based on blood transport rate attenuation data to obtain blood fluidity pressure intensity data; Performing omnidirectional stress tensor calculation on the blood fluidity pressure intensity data to obtain omnidirectional stress tensor data; Based on the omnidirectional stress tensor data, stress spatial distribution fluctuation analysis is performed to obtain stress spatial distribution fluctuation data; The spatial potential energy relationship of stress spatial distribution fluctuation data is quantified according to the Bernoulli equation to obtain stress spatial potential energy relationship data; The pressure increment azimuth decomposition calculation is performed based on the stress space potential energy relationship data and the stress space distribution fluctuation data to obtain the blood pressure increment azimuth decomposition data.

5. The method for constructing a risk assessment model for myocardial infarction patients according to claim 1, characterized in that: Step S32 includes the following steps: Step S321: calculating the phase offset of the blocked feedback ECG signal phase offset data to obtain the blocked ECG signal phase offset data; Step S322: performing nonlinear incremental calculation on the blocked feedback ECG signal phase offset data according to the blocked ECG signal phase offset data to obtain phase offset nonlinear incremental data; Step S323: performing spatial increment trend analysis on the phase offset data of the blocked feedback ECG signal according to the phase offset nonlinear increment data to obtain phase offset spatial increment trend data; Step S324: performing spatial interpolation processing on the phase offset data of the blocked feedback ECG signal based on the phase offset spatial increment trend data to obtain signal offset spatial interpolation data.

6. The method for constructing a risk assessment model for myocardial infarction patients according to claim 1, characterized in that: Step S34 includes the following steps: Step S341: dividing the blocking feedback signal phase offset compensation data into a data set to obtain a blocking feedback signal phase offset compensation test set and a blocking feedback signal phase offset compensation training set; Step S342: performing feedback signal similarity feature selection on the blocking feedback signal phase offset compensation training set to obtain feedback signal similarity feature data; Step S343: using a random forest algorithm and performing similar structure incremental learning on the blocking feedback signal phase offset compensation training set based on the similar feature data of the feedback signal to obtain similar structure incremental data; Step S344: constructing an initial risk assessment model for myocardial infarction patients based on the similar structure incremental data to obtain an initial risk assessment model for myocardial infarction patients; Step S345: performing a model test on the initial myocardial infarction patient risk assessment model based on the blocking feedback signal phase offset compensation test set to obtain a myocardial infarction patient risk assessment model.

7. A system for constructing a risk assessment model for myocardial infarction patients, characterized in that: A method for constructing a risk assessment model for myocardial infarction patients according to claim 1, wherein the system for constructing the risk assessment model for myocardial infarction patients comprises: The angiography 3D model construction module is used to collect vital sign data and coronary angiography images of myocardial infarction patients through the medical information system, thereby obtaining vital sign records of myocardial infarction patients and coronary angiography images of myocardial infarction patients, respectively; and construct an angiography multi-view 3D model based on the coronary angiography images of myocardial infarction patients, thereby obtaining an angiography multi-view 3D model. The ECG signal pre-gradient feedback recognition module is used to perform ECG signal-vascular motion mapping processing on the multi-view 3D model of arteriography to obtain ECG signal-vascular motion mapping data; perform vascular blockage increment calculation on the ECG signal-vascular motion mapping data to obtain vascular blockage increment data; and perform ECG signal pre-gradient feedback recognition on the vascular blockage increment data based on the ECG signal-vascular motion mapping data to obtain ECG pre-gradient feedback data; The myocardial infarction patient risk assessment model construction module is used to perform ECG signal phase offset analysis on the blocked ECG pre-gradient feedback data to obtain blocked feedback ECG signal phase offset data; a myocardial infarction patient risk assessment model is constructed using the blocked feedback ECG signal phase offset data using a random forest algorithm to obtain a myocardial infarction patient risk assessment model; The automated monitoring operation module is used to design automated monitoring firmware for the risk assessment model for myocardial infarction patients, obtain the monitoring firmware for the risk assessment model for myocardial infarction patients, and send the monitoring firmware for the risk assessment model for myocardial infarction patients to the cloud platform to perform risk assessment for myocardial infarction patients.

Images