A dynamic prediction system for long-term adverse cardiovascular events after stent placement

Abstract

This invention discloses a dynamic prediction system for long-term adverse cardiovascular event risk after stent implantation, belonging to the field of personal health risk assessment technology. It includes: a cardiovascular assessment module for determining the target cardiovascular adverse event presentation coefficient based on the target patient's first chest pain rating and real-time heart rate after stent implantation; a cardiac function assessment module for determining the emphasis of cardiac function analysis at each stage based on the target patient's physical examination data after stent implantation; a probability determination module for determining the probability of the first adverse event based on the target cardiovascular adverse event presentation coefficient and the emphasis of each cardiac function analysis; and an indicator determination module for comparing the target patient's first adverse event probability with the second adverse event probability of each monitored patient to obtain the target cardiovascular risk warning indicator for the target patient. This invention can improve the accuracy of long-term risk prediction for adverse cardiovascular events.

Description

Technical Field

[0001] This invention relates to the field of personal health risk assessment technology, specifically to a dynamic prediction system for long-term adverse cardiovascular event risk after stent placement. Background Technology

[0002] Percutaneous coronary intervention (PCI) is a core clinical treatment for coronary artery disease, rapidly opening blocked blood vessels and restoring myocardial perfusion. However, adverse cardiovascular events such as in-stent restenosis, thrombosis, and heart failure not only directly threaten patients' lives and health but also significantly reduce their quality of life, increase long-term medical and economic burdens, and accelerate the decline of cardiovascular function after the procedure. Therefore, long-term risk monitoring and dynamic early warning for patients after PCI are of significant clinical importance.

[0003] Currently, traditional methods for predicting postoperative risks in stent implantation patients are mostly static assessment models, which mainly rely on clinical indicators (such as blood pressure, blood lipids, blood glucose, etc.) at a fixed time point after the operation for risk stratification analysis.

[0004] However, existing static assessments lack real-time responsiveness to cardiovascular events and do not fully integrate immediate vital signs such as postoperative heart rate, resulting in poor accuracy in predicting the long-term risk of adverse cardiovascular events. Summary of the Invention

[0005] This invention provides a dynamic prediction system for the long-term adverse cardiovascular event risk after stent placement, which can improve the accuracy of long-term risk prediction for adverse cardiovascular events.

[0006] A first aspect of the present invention provides a dynamic prediction system for the risk of long-term adverse cardiovascular events after stent placement, comprising: The cardiovascular assessment module is used to determine the target cardiovascular adverse presentation coefficient for the target patient based on the first chest pain rating and real-time heart rate after stent implantation. The target cardiovascular adverse presentation coefficient is used to characterize the immediate cardiovascular status of the target patient. The cardiac function assessment module is used to determine the emphasis of cardiac function analysis at each stage based on the physical examination data of the target patient at each stage after stent implantation; the emphasis of cardiac function analysis is used to characterize the overall status of the target patient's cardiac function. The probability determination module is used to determine the probability of the first adverse event in a target patient based on the target cardiovascular adverse event presentation coefficient and the emphasis of each cardiac function analysis. The indicator determination module is used to compare the probability of the first adverse event in the target patient with the probability of the second adverse event in each monitored patient to obtain the target cardiovascular risk warning indicator for the target patient.

[0007] Furthermore, the present invention also proposes that the cardiovascular assessment module includes: The cardiovascular assessment submodule is used to determine the first cardiovascular adverse presentation coefficient of the target patient based on the first chest pain rating and real-time heart rate after stent implantation; the first cardiovascular adverse presentation coefficient is the cardiovascular adverse presentation coefficient of the target patient after stent implantation. The deviation assessment submodule is used to subtract the baseline value of the cardiovascular adverse presentation coefficient from the first cardiovascular adverse presentation coefficient of the target patient to obtain the deviation value of the cardiovascular adverse presentation coefficient; the baseline value of the cardiovascular adverse presentation coefficient is the average value of the second cardiovascular adverse presentation coefficient of the target patient multiple times before stent implantation; the second cardiovascular adverse presentation coefficient is the cardiovascular adverse presentation coefficient of the target patient before stent implantation. The evaluation correction submodule is used to correct the first cardiovascular adverse presentation coefficient based on the deviation value of the cardiovascular adverse presentation coefficient, so as to obtain the target cardiovascular adverse presentation coefficient for the target patient.

[0008] Furthermore, the present invention also proposes that the cardiovascular assessment submodule includes: The chest pain assessment unit is used to determine the intensity of the chest pain response of the target patient based on the frequency of chest pain and the corresponding first chest pain rating within a preset time after stent implantation. The heart rate assessment unit is used to determine the cardiac factor prevalence of the target patient based on the real-time heart rate of the target patient within a preset time after stent implantation; the cardiac factor prevalence is used to characterize the degree of excellence of the heart rate status. The cardiovascular assessment unit is used to determine the first cardiovascular adverse presentation coefficient of the target patient based on the intensity of chest pain response and the prevalence of cardiac factors.

[0009] Furthermore, the present invention also proposes that the chest pain assessment unit is used for: The mean of each first chest pain rating of the target patients was obtained by averaging the first chest pain ratings. The mean of the second chest pain ratings for each monitored patient was calculated. The chest pain rating score for the target patient is obtained by multiplying the ratio of the mean of the first chest pain rating to the mean of the second chest pain rating by the frequency of chest pain. The chest pain rating values ​​of the target patients were normalized to obtain the intensity of their chest pain response.

[0010] Furthermore, the present invention also proposes that the heart rate assessment unit is used for: Based on the target patient’s real-time heart rate within a preset time after stent implantation, extract abnormal heart rate segments where the real-time heart rate is lower than the first heart rate threshold or higher than the second heart rate threshold; the second heart rate threshold is greater than the first heart rate threshold. For each abnormal heart rate segment, the absolute value of the difference between the average heart rate of the first heart rate within the abnormal heart rate segment and the average heart rate of the second heart rate within a preset duration is taken to obtain the heart rate deviation value of each abnormal heart rate segment. The mean heart rate deviation values ​​of each abnormal heart rate segment were averaged to obtain the mean heart rate deviation. The priority of cardiac factors for the target patient is determined based on the number of abnormal segments of heart rate and the mean heart rate deviation.

[0011] Furthermore, the present invention also proposes that the physical examination data include the mean ST segment value of electrocardiogram, platelet count, left ventricular ejection fraction, and coronary angiography stenosis degree; The cardiac function assessment module includes: The thrombosis assessment submodule is used to determine the thrombosis trend of the target patient at the target stage based on the mean ST segment value and platelet count of the electrocardiogram at the target stage after stent implantation. The cardiac function assessment submodule is used to determine the emphasis of cardiac function analysis in the target patient at the target stage based on thrombosis trend, left ventricular ejection fraction, and coronary angiography stenosis.

[0012] Furthermore, the present invention also proposes that the thrombosis assessment submodule is used for: The absolute value of the difference between the mean ST segment value of the target patient's ECG at the target stage and the maximum ST segment value of the ECG in each monitored patient is divided by the maximum ST segment value of the ECG to obtain the ECG ST segment variability. The platelet count variability is obtained by dividing the absolute value of the difference between the platelet count of the target patient at the target stage and the maximum platelet count among all monitored patients by the maximum platelet count. Based on the ST segment variability of electrocardiogram and the platelet count variability, the thrombosis trend of the target patient at the target stage is determined.

[0013] Furthermore, the present invention also proposes that the probability determination module is used for: The emphasis of each cardiac function analysis was fitted with a straight line in chronological order to obtain the target fitted straight line; Based on the slope of the target fitted line, the target cardiac function analysis emphasis, and the target cardiovascular adverse event presentation coefficient, the probability of the first adverse event for the target patient is determined; the target cardiac function analysis emphasis is the cardiac function analysis emphasis closest to the current time among all cardiac function analysis emphasis.

[0014] Furthermore, the present invention also proposes that the index determination module is used for: The mean of the second adverse event probability for each monitored patient was calculated. The difference between the probability of the first adverse event and the mean probability of adverse events in the target patients is normalized to obtain the target cardiovascular risk warning index for the target patients.

[0015] Furthermore, the present invention also proposes that, after comparing the probability of a first adverse event in the target patient with the probability of a second adverse event in each monitored patient to obtain the target cardiovascular risk warning index for the target patient, the invention further includes: The indicator analysis module is used to compare the target cardiovascular risk warning indicators of the target patient with the first warning indicator threshold and the second warning indicator threshold to obtain the indicator comparison results; the second warning indicator threshold is greater than the first warning indicator threshold. The indicator analysis module is also used to initiate a first warning reminder to the target patient's communication terminal in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is greater than the second warning indicator threshold; the first warning reminder is the highest level of warning reminder; The indicator analysis module is also used to continue risk prediction for the target patient in response to the indicator comparison results indicating that the target cardiovascular risk warning indicator is less than the first warning indicator threshold. The indicator analysis module is also used to initiate a second warning reminder corresponding to the target cardiovascular risk warning indicator to the target patient's communication terminal in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is not less than the first warning indicator threshold and not greater than the second warning indicator threshold.

[0016] The present invention has the following beneficial effects: The long-term adverse cardiovascular event risk dynamic prediction system provided in this invention acquires the first chest pain rating and real-time heart rate of the target patient after surgery through a cardiovascular assessment module. This fully integrates and utilizes the patient's immediate postoperative vital signs information, enabling dynamic capture and quantitative characterization of the immediate cardiovascular status, thus overcoming the lack of real-time responsiveness in existing technologies. Simultaneously, the cardiac function assessment module determines the emphasis of cardiac function analysis at each stage based on the patient's postoperative physical examination data, achieving long-term, phased tracking and quantitative analysis of the overall cardiac function status, overcoming the limitations of traditional methods that only use data from a single fixed time point for assessment. Furthermore, the probability determination module integrates the immediate cardiovascular status and phased cardiac function status to obtain a first adverse event probability that more closely reflects the patient's actual condition. This probability is then compared with the monitored patient population by the indicator determination module, ultimately outputting a target cardiovascular risk warning indicator. Thus, long-term, dynamic, and multi-dimensional risk prediction of adverse cardiovascular events is achieved, improving the accuracy of long-term risk prediction for adverse cardiovascular events. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of the structure of a dynamic prediction system for long-term adverse cardiovascular events after stent placement, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a cardiovascular assessment module provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a cardiovascular assessment submodule provided in one embodiment of the present invention; Figure 4 This is a schematic diagram of the cardiac function assessment module provided in one embodiment of the present invention. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a dynamic prediction system for long-term adverse cardiovascular events after stent placement, as proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] In traditional stent implantation postoperative risk prediction, the static assessment model relies solely on clinical indicators at fixed postoperative time points for risk stratification analysis. It lacks a dynamic integration mechanism for real-time vital signs such as postoperative heart rate, resulting in the system's inability to respond to cardiovascular event risks in real time, thus reducing the accuracy of long-term risk prediction. The essence of this problem lies in the fact that static assessment fails to establish a dynamic correlation model between real-time vital sign data and cardiovascular status, causing the risk warning mechanism to deviate from the actual physiological changes in patients, thereby affecting the timeliness and reliability of risk prediction.

[0022] For example, in a follow-up scenario in the cardiovascular department of a hospital, a target patient reported sudden chest pain on the third day after stent implantation. At the same time, their wearable device monitored a real-time heart rate that was consistently higher than a preset threshold. However, the existing system only uses static blood pressure and blood lipid data from one week after the surgery for evaluation, without incorporating chest pain rating and real-time heart rate changes into the analysis process. As a result, the system cannot identify the deteriorating trend of the patient's immediate cardiovascular status, which in turn causes the risk warning mechanism to fail and delays the timing of clinical intervention.

[0023] If the above problems are not addressed, cardiovascular event risk monitoring will be unable to achieve dynamic early warning, and patients may continue to be exposed to unidentified high-risk conditions, increasing the probability of adverse cardiovascular events such as in-stent restenosis and thrombosis. At the same time, it will lead to a decrease in the efficiency of medical resource allocation and an acceleration of the postoperative cardiovascular function decline. The continued existence of this technical defect will directly weaken the clinical application value of the risk prediction system.

[0024] In this regard, such as Figure 1 As shown in the figure, the present invention proposes a structural schematic diagram of a dynamic prediction system for long-term adverse cardiovascular events after stent placement. The dynamic prediction system 100 for long-term adverse cardiovascular events after stent placement includes a cardiovascular assessment module 110, a cardiac function assessment module 120, a probability determination module 130, and an indicator determination module 140.

[0025] The cardiovascular assessment module 110 is used to determine the target cardiovascular adverse presentation coefficient of the target patient based on the first chest pain rating and real-time heart rate of the target patient after stent implantation; the target cardiovascular adverse presentation coefficient is used to characterize the immediate cardiovascular status of the target patient. The cardiac function assessment module 120 is used to determine the emphasis of cardiac function analysis at each stage of the target patient's physical examination data after stent implantation; the emphasis of cardiac function analysis is used to characterize the overall status of the target patient's cardiac function. The probability determination module 130 is used to determine the probability of the first adverse event in the target patient based on the target cardiovascular adverse event presentation coefficient and the emphasis of each cardiac function analysis. The indicator determination module 140 is used to compare the probability of the first adverse event of the target patient with the probability of the second adverse event of each monitored patient to obtain the target cardiovascular risk warning indicator for the target patient.

[0026] For ease of understanding, the following explains some key terms in this embodiment: The cardiovascular assessment module 110 is a functional unit in the system responsible for collecting and processing data related to the patient's real-time cardiovascular status. This module primarily quantifies the degree of current cardiovascular adverse events based on the patient's real-time physiological indicators, such as chest pain rating and heart rate.

[0027] The target patient refers to the specific individual for whom the system focuses and predicts risk. This patient is a subject who has undergone stent implantation, and their data are used to assess and predict their risk of adverse cardiovascular events.

[0028] The Primary Chest Pain Rating Scale (PCR) is a quantitative assessment of chest pain symptoms experienced by target patients after stent implantation. This rating reflects characteristics such as the frequency, intensity, and duration of chest pain, and is an important basis for assessing immediate cardiovascular status.

[0029] Real-time heart rate refers to the frequency of the heartbeats of a target patient at a specific point in time or over a specific duration. This data directly reflects the patient's cardiac activity and is a key physiological parameter in cardiovascular assessment.

[0030] The target cardiovascular adverse event presentation coefficient is a comprehensive value calculated by the cardiovascular assessment module 110, used to characterize the current immediate cardiovascular status of the target patient. The value of this coefficient is associated with the immediate risk of the patient experiencing adverse cardiovascular events.

[0031] The cardiac function assessment module 120 is a functional unit in the system responsible for collecting and processing data related to the patient's overall cardiac function. This module primarily assesses the long-term trend and stability of the patient's cardiac function based on physical examination data from different stages.

[0032] Physical examination data refers to the collection of data obtained from physical examinations performed on target patients at different time points after stent implantation. This data may include electrocardiograms, complete blood counts, imaging results, etc., and is used to comprehensively assess the patient's cardiac function.

[0033] The cardiac function analysis emphasis score is a numerical value calculated by the cardiac function assessment module 120, used to characterize the overall cardiac function status of the target patient. This cardiac function analysis emphasis score reflects the health level or potential risk of the patient's cardiac function at a specific stage.

[0034] The probability determination module 130 is a functional unit in the system responsible for integrating immediate cardiovascular status and long-term cardiac function to calculate the probability of a patient experiencing adverse events. This module synthesizes assessment results from different dimensions to determine the likelihood of a patient experiencing adverse events in the future.

[0035] The first adverse event probability refers to the predicted probability of an adverse cardiovascular event occurring in the target patient, calculated by the probability determination module 130. This probability is a quantitative prediction of the target patient's future risk by the system.

[0036] The indicator determination module 140 is a functional unit in the system responsible for comparing the predicted probability of the target patient with that of the reference group to generate risk warning indicators. This module provides personalized risk warning information to patients through comparative analysis.

[0037] The monitored patients refer to a reference group with similar clinical characteristics or backgrounds to the target patients, who are included in the system for data collection and analysis. The probability of a second adverse event in this group is used for comparison with the probability in the target patients.

[0038] The second adverse event probability refers to the predicted probability of an adverse cardiovascular event occurring in a monitored patient population. This probability can be the probability of an individual monitored patient or the average probability of the monitored patient population, serving as a reference for risk assessment of the target patients.

[0039] The target cardiovascular risk warning index is a comprehensive indicator calculated by the indicator determination module 140, used to intuitively reflect the risk level of adverse cardiovascular events in target patients. This indicator can be used to guide clinical decision-making and patient management.

[0040] The cardiovascular assessment module 110 is configured to determine a target cardiovascular adverse presentation coefficient for a target patient based on the patient's first chest pain rating and real-time heart rate after stent implantation. This target cardiovascular adverse presentation coefficient characterizes the patient's immediate cardiovascular status. For example, the cardiovascular assessment module 110 can receive the first chest pain rating reported by the patient via questionnaire or verbal communication, and real-time heart rate data acquired through a wearable device or medical monitor. The module can then use a pre-defined weighted average algorithm to simply combine the chest pain rating and real-time heart rate; for example, multiplying the chest pain rating by a weighting factor and multiplying the degree of abnormality in the real-time heart rate (e.g., the extent to which it exceeds the normal range) by another weighting factor, and then summing the two to obtain the target cardiovascular adverse presentation coefficient. A higher target cardiovascular adverse presentation coefficient indicates a more unstable or adverse tendency in the patient's current immediate cardiovascular status.

[0041] The cardiac function assessment module 120 is configured to determine the cardiac function analysis emphasis for each stage of the target patient's post-stent implantation physical examination data. This cardiac function analysis emphasis characterizes the overall cardiac function status of the target patient. Specifically, the physical examination data may include the results of routine examinations performed on the patient at different time points, such as basic physiological indicators like blood pressure, blood lipids, and blood glucose. The cardiac function assessment module 120 can perform simple statistical analysis on this physical examination data, for example, calculating the average value or deviation from the normal range of various indicators, and then simply summing or averaging these analysis results to generate a comprehensive cardiac function analysis emphasis that reflects the patient's cardiac function health level at a specific stage.

[0042] The probability determination module 130 is configured to determine the probability of a first adverse event for the target patient based on the target cardiovascular adverse event presentation coefficient and the emphasis of each cardiac function analysis. This module integrates the assessment results of immediate cardiovascular status with the assessment results of long-term cardiac function. For example, the probability determination module 130 can employ a simple linear model to weight and sum the target cardiovascular adverse event presentation coefficient with the emphasis of one or more recent cardiac function analyses, and then map it to a probability value between 0 and 1 using a pre-defined transformation function (e.g., the sigmoid function), thereby obtaining the probability of the target patient's first adverse event. This probability value directly quantifies the likelihood of the target patient experiencing an adverse cardiovascular event in the future.

[0043] The indicator determination module 140 is configured to compare the probability of a first adverse event in the target patient with the probability of a second adverse event in each monitored patient to obtain a target cardiovascular risk warning indicator for the target patient. This module aims to provide a relative risk assessment. Specifically, the indicator determination module 140 can simply calculate the difference between the probability of a first adverse event in the target patient and the probability of a second adverse event in one or more monitored patients. For example, the probability of the target patient can be directly subtracted from the probability of a monitored patient representing an average risk level to obtain a difference as the target cardiovascular risk warning indicator. The magnitude of this indicator can intuitively reflect the degree of deviation of the target patient's risk from the reference group.

[0044] The following example will provide a more detailed explanation of the above technical solution: Suppose user A needs dynamic prediction of their long-term adverse cardiovascular event risk after stent implantation. The system first assesses user A's immediate cardiovascular status through a cardiovascular assessment module 110. Specifically, the system receives user A's chest pain rating reported on a post-operative day (e.g., a rating of 3 on a simple pain scale, indicating mild chest pain), and simultaneously acquires real-time heart rate data monitored by user A via a smartwatch (e.g., an average heart rate of 85 beats / minute, slightly higher than their baseline heart rate). The cardiovascular assessment module 110 performs preliminary processing on this data, such as weighting the chest pain rating with the degree of heart rate abnormality (e.g., the duration or magnitude of heart rate exceeding the normal range), thereby calculating user A's current target adverse cardiovascular event presentation coefficient, which reflects user A's current immediate cardiovascular status.

[0045] Simultaneously, the cardiac function assessment module 120 will evaluate the overall status of user A's cardiac function based on user A's physical examination data at different stages after stent implantation. For example, the system can obtain user A's routine physical examination reports at 1 month, 3 months, and 6 months post-surgery. These reports may include basic indicators such as blood pressure, blood lipids, and blood sugar. The cardiac function assessment module 120 analyzes this historical physical examination data, for example, calculating the changing trends of various indicators over time and synthesizing them into a cardiac function analysis emphasis score. Suppose that user A's cardiac function analysis emphasis scores are 0.4, 0.5, and 0.6 at 1 month, 3 months, and 6 months, respectively, indicating that their cardiac function may be gradually worsening.

[0046] Subsequently, the probability determination module 130 integrates the aforementioned assessment results. This module receives user A's current target cardiovascular adverse event presentation coefficient (e.g., 0.7, indicating immediate poor condition) and the emphasis of their cardiac function analysis at each stage (e.g., 0.6 in the most recent analysis). The probability determination module 130 inputs this data into a preset risk prediction model, which can be a machine learning-based model or a statistical regression model. Using this model, the system calculates the probability of user A's first adverse event, quantifying the likelihood of user A experiencing adverse cardiovascular events such as in-stent restenosis, thrombosis, or heart failure within a future period. For example, the calculated probability of the first adverse event is 0.15.

[0047] Finally, the indicator determination module 140 compares the probability of user A's first adverse event with the probability of the monitored patients' second adverse event. Assuming the system maintains a database of monitored patients with similar age, gender, and medical history to user A, the probability of the second adverse event for these monitored patients can be calculated. The indicator determination module 140 compares user A's 0.15 with the monitored patients' 0.10, for example, calculating the difference or ratio between the two, to obtain user A's target cardiovascular risk warning indicator. This indicator can visually show whether user A's risk level is higher or lower than that of similar patients, thus providing decision support for doctors. For example, if the indicator shows a higher risk, the doctor may recommend that user A undergo more frequent follow-up examinations or adjust the treatment plan.

[0048] Based on the above examples, the dynamic prediction system 100 for long-term adverse cardiovascular events after stent placement proposed in this invention demonstrates significant technical contributions. Existing risk prediction methods often employ static assessment models, relying solely on clinical indicators at a fixed post-operative time point for analysis. This results in insufficient real-time responsiveness to cardiovascular events and fails to fully integrate the patient's immediate vital signs, thus affecting the accuracy of long-term risk prediction.

[0049] The system of this invention, by introducing a cardiovascular assessment module 110, can dynamically determine the target cardiovascular adverse presentation coefficient based on the target patient's real-time physiological data, such as real-time heart rate and first chest pain rating, thereby capturing changes in the patient's cardiovascular status in real time. For example, in the case of user A, the system not only considers their historical physical examination data, but more importantly, it can respond instantly to the chest pain and real-time heart rate data reported by user A. This makes the assessment of the patient's immediate risk more timely and accurate, overcoming the limitations of traditional static assessment.

[0050] Furthermore, the cardiac function assessment module 120 of the present invention can determine the emphasis of cardiac function analysis based on the patient's physical examination data at various stages, thus characterizing the overall status of cardiac function. Unlike traditional methods that rely solely on data from a single point in time, this system can integrate multi-dimensional physical examination data from patients at different time points to form a comprehensive view of the long-term trend of cardiac function.

[0051] By organically combining immediate cardiovascular status (target cardiovascular adverse event presentation coefficient) with long-term cardiac function status (cardiac function analysis emphasis) through the probability determination module 130, this system can generate a more comprehensive and accurate first adverse event probability. This multi-dimensional and dynamic data integration method improves the accuracy and reliability of prediction and avoids the bias that may be caused by a single indicator or static assessment.

[0052] Ultimately, the indicator determination module 140 generates a target cardiovascular risk warning indicator by comparing the probability of a first adverse event in the target patient with the probability of a second adverse event in the monitored patients. This relativistic risk assessment approach allows physicians to more intuitively understand where a patient's risk level stands within their peer group, thereby enabling them to make more targeted clinical decisions.

[0053] The long-term adverse cardiovascular event risk dynamic prediction system 100 provided in this embodiment obtains the first chest pain rating and real-time heart rate of the target patient after surgery through the cardiovascular assessment module 110. It can fully integrate and utilize the patient's immediate postoperative vital signs information to achieve dynamic capture and quantitative characterization of the immediate cardiovascular status, making up for the lack of real-time responsiveness in existing technologies. At the same time, the cardiac function assessment module 120 determines the emphasis of cardiac function analysis at each stage based on the patient's postoperative physical examination data, realizing long-term, staged tracking and quantitative analysis of the overall cardiac function status, breaking through the limitation of traditional methods that only use data from a single fixed time point for assessment. Then, the probability determination module 130 integrates the immediate cardiovascular status and staged cardiac function status to obtain the first adverse event probability that is more in line with the patient's actual condition. The indicator determination module 140 compares it with the monitored patient group and finally outputs the target cardiovascular risk warning indicator. In this way, long-term, dynamic, and multi-dimensional risk prediction of adverse cardiovascular events is realized, which can improve the accuracy of long-term risk prediction of adverse cardiovascular events.

[0054] In some embodiments of the present invention described above, the system directly determines the target cardiovascular adverse presentation coefficient of the target patient based on the target patient's first chest pain rating and real-time heart rate after stent implantation via the cardiovascular assessment module 110. However, this direct assessment method may fail to fully consider the physiological differences between individual patients and their preoperative baseline status, resulting in the determined target cardiovascular adverse presentation coefficient failing to accurately reflect the patient's immediate cardiovascular status, thereby affecting the accuracy and personalization of subsequent risk prediction.

[0055] In this regard, such as Figure 2 As shown, the present invention further proposes that the cardiovascular assessment module 110 includes: The cardiovascular assessment submodule 111 is used to determine the first cardiovascular adverse presentation coefficient of the target patient based on the first chest pain rating and real-time heart rate after stent implantation; the first cardiovascular adverse presentation coefficient is the cardiovascular adverse presentation coefficient of the target patient after stent implantation. The deviation assessment submodule 112 is used to subtract the baseline value of the cardiovascular adverse presentation coefficient from the first cardiovascular adverse presentation coefficient of the target patient to obtain the deviation value of the cardiovascular adverse presentation coefficient; the baseline value of the cardiovascular adverse presentation coefficient is the average value of the second cardiovascular adverse presentation coefficient of the target patient multiple times before stent implantation; the second cardiovascular adverse presentation coefficient is the cardiovascular adverse presentation coefficient of the target patient before stent implantation. The evaluation correction submodule 113 is used to correct the first cardiovascular adverse presentation coefficient based on the deviation value of the cardiovascular adverse presentation coefficient, so as to obtain the target cardiovascular adverse presentation coefficient of the target patient.

[0056] In this embodiment, the cardiovascular assessment submodule 111 performs a preliminary assessment based on the target patient's immediate physiological data after stent implantation, namely the first chest pain rating and real-time heart rate, to determine an initial first cardiovascular adverse presentation coefficient. This coefficient serves as the basis for subsequent corrections and can be determined either by using a preset weighted algorithm model to comprehensively calculate different levels of first chest pain ratings and real-time heart rate values, or by consulting a pre-established lookup table to map a specific combination of chest pain rating and heart rate to a preliminary coefficient.

[0057] The deviation assessment submodule 112 introduces a personalized assessment dimension. It quantifies the deviation between the current cardiovascular status and the patient's preoperative health level by comparing the initially determined first cardiovascular adverse presentation coefficient with the target patient's own historical baseline data. The baseline value of the cardiovascular adverse presentation coefficient is the mean of the target patient's multiple second cardiovascular adverse presentation coefficients before stent implantation (e.g., during the surgical observation period), which provides a personalized reference point for assessment.

[0058] Based on this, the evaluation correction submodule 113 uses the deviation value obtained from the deviation evaluation submodule 112 to finely adjust the preliminary first cardiovascular adverse presentation coefficient. This correction mechanism ensures that the final target cardiovascular adverse presentation coefficient not only reflects the patient's current immediate physiological condition but also incorporates their individualized historical health background, thus making it more representative and accurate.

[0059] The present invention subdivides the cardiovascular assessment module into a cardiovascular assessment submodule 111, a deviation assessment submodule 112, and an assessment correction submodule 113, forming a progressively refined and personalized assessment process. First, the cardiovascular assessment submodule 111 provides a preliminary first cardiovascular adverse presentation coefficient based on the patient's immediate postoperative physiological indicators. Then, the deviation assessment submodule 112 compares this preliminary coefficient with the patient's own preoperative cardiovascular adverse presentation coefficient baseline value to calculate a deviation value reflecting individual differences. Finally, the assessment correction submodule 113 uses this deviation value to correct the preliminary coefficient, thereby obtaining a more accurate target cardiovascular adverse presentation coefficient that fully considers the patient's individual characteristics. This series of synergistic modules ensures that the assessment of the target patient's immediate cardiovascular status is no longer a simple absolute value judgment, but rather a dynamic comparison and correction based on their own historical data, significantly improving the personalization and accuracy of the assessment and providing a more reliable input for subsequent risk prediction.

[0060] As an example, the target cardiovascular adverse presentation coefficient for the target patient can be determined using the following formula 1: Formula 1 In formula 1, The target cardiovascular adverse presentation coefficient is used to characterize the target patient. The bias value of the cardiovascular adverse presentation coefficient is used to characterize the target patient; th is used to characterize the hyperbolic tangent function operation. The first cardiovascular adverse presentation coefficient is used to characterize the target patient.

[0061] Specifically, if the target patient's current real-time cardiovascular adverse symptoms are more significantly worse than their preoperative cardiovascular adverse symptoms, it reflects a higher level of attention to the target patient's cardiovascular health, and the cardiovascular adverse symptoms presentation coefficient should be increased accordingly; conversely, it should be reduced.

[0062] Through the aforementioned technical solutions, the system can fully consider the individual physiological differences and preoperative baseline status of target patients, enabling a more accurate and personalized assessment of their immediate cardiovascular status after stent implantation. This allows the determined target cardiovascular adverse event presentation coefficient to more realistically reflect the patient's actual situation, thereby significantly improving the accuracy and reliability of dynamic prediction of long-term adverse cardiovascular event risks after stent implantation, and providing a more valuable reference for patient health management.

[0063] In some of the embodiments of the present invention described above, if the original first chest pain rating and real-time heart rate data are directly used to calculate the first cardiovascular adverse presentation coefficient, it may not be able to fully capture the complexity and dynamic changes of the patient's cardiovascular status, resulting in insufficient accuracy and robustness of the assessment results, thereby affecting the accuracy of subsequent risk prediction.

[0064] In this regard, such as Figure 3 As shown, the present invention further proposes that the cardiovascular assessment submodule 111 includes: Chest pain assessment unit 111a is used to determine the intensity of chest pain response of the target patient based on the frequency of chest pain and the corresponding first chest pain rating within a preset time after stent implantation. The heart rate assessment unit 111b is used to determine the cardiac factor eugenicity of the target patient based on the real-time heart rate of the target patient within a preset time after stent implantation; the cardiac factor eugenicity is used to characterize the degree of excellence of the heart rate status. The cardiovascular assessment unit 111c is used to determine the first cardiovascular adverse presentation coefficient of the target patient based on the intensity of the target patient's chest pain response and the priority of cardiac factors.

[0065] In this embodiment, the chest pain assessment unit 111a is a functional module whose function is to quantify the patient's chest pain status based on the frequency of chest pain occurrence and the severity of each episode (first chest pain rating) within a specific time period after stent implantation, and output a comprehensive index, namely, chest pain response intensity. This unit can be implemented by software algorithms, such as processing the input chest pain data through a dedicated calculation program segment. Chest pain frequency refers to the number of times the target patient experiences chest pain within a preset time period after stent implantation, and this data can be obtained through patient self-reporting, medical records, or wearable device monitoring. The first chest pain rating refers to the quantitative assessment of the severity of each episode of chest pain in the target patient, usually using a preset scoring standard, such as a 0-10 scale, with higher scores indicating more severe chest pain. This rating can be subjectively assessed by the patient based on their own feelings, or objectively assessed by medical staff based on clinical manifestations. Chest pain response intensity is a comprehensive indicator used to characterize the overall severity and frequency of chest pain in a target patient within a preset time period. It combines the number of chest pains with the rating of each chest pain to provide a quantitative value that reflects the patient's cardiovascular status more effectively than a single rating or frequency.

[0066] The heart rate assessment unit 111b is another functional module. Its function is to assess the patient's heart rate status based on real-time heart rate data within a specific time period after stent implantation and output an index characterizing the excellence of the heart rate status, namely, the cardiac factor eugenics. This unit can be implemented by software algorithms, such as through a data analysis program to process continuous heart rate data. The preset duration refers to the time window used to collect chest pain frequency, first chest pain rating, and real-time heart rate data. This duration can be set according to clinical needs and data collection frequency, for example, it can be 24 hours. Real-time heart rate refers to the heart rate data of the target patient continuously or periodically measured within the preset duration after stent implantation. This data can be acquired in real time through electrocardiogram monitoring devices, wearable heart rate sensors, etc. The cardiac factor eugenics is a comprehensive index used to characterize the excellence of the target patient's heart rate status. It not only considers the average heart rate but may also consider heart rate variability, the frequency of abnormal heart rate events, etc., to provide a more comprehensive assessment of heart rate health.

[0067] The cardiovascular assessment unit 111c is the core functional module. Its function is to calculate the first cardiovascular adverse presentation coefficient of the target patient by integrating the chest pain response intensity output by the chest pain assessment unit 111a and the cardiac factor dominance output by the heart rate assessment unit 111b. This unit can be implemented by software algorithms, such as through a multivariate regression model, a neural network model, or a rule-based expert system to fuse these two indicators. The first cardiovascular adverse presentation coefficient is a quantitative indicator used to initially characterize the immediate cardiovascular status of the target patient. It integrates the patient's chest pain and heart rate status, providing basic data for subsequent bias assessment and correction.

[0068] In this invention, to more precisely assess the immediate cardiovascular status of the target patient, the cardiovascular assessment submodule 111 is designed to include a chest pain assessment unit 111a, a heart rate assessment unit 111b, and a cardiovascular assessment unit 111c. Specifically, the chest pain assessment unit 111a first receives the frequency of chest pain and the corresponding first chest pain rating of the target patient within a preset time period after stent implantation. This unit processes this raw chest pain data, for example, by combining the frequency of chest pain occurrences and the severity of each occurrence, to quantify and generate a chest pain response intensity that comprehensively reflects the patient's chest pain status. Simultaneously, the heart rate assessment unit 111b receives the real-time heart rate data of the target patient within the same preset time period. This unit analyzes the real-time heart rate data, for example, by assessing the stability of the heart rate and whether there are abnormal fluctuations, thereby determining a cardiac factor dominance degree that characterizes the excellence of the heart rate status. Subsequently, the cardiovascular assessment unit 111c uses the chest pain response intensity output by the chest pain assessment unit 111a and the cardiac factor dominance degree output by the heart rate assessment unit 111b as inputs. This unit uses a pre-defined algorithm or model to fuse and calculate two independent but both key indicators reflecting the immediate cardiovascular status, ultimately determining the first cardiovascular adverse presentation coefficient for the target patient. This step-by-step, multi-dimensional assessment approach makes the determination of the first cardiovascular adverse presentation coefficient more comprehensive and accurate, avoiding information loss or assessment bias that may result from directly using raw data, thereby improving the overall system's accuracy in determining the target cardiovascular adverse presentation coefficient.

[0069] As an example, the first cardiovascular adverse presentation coefficient of the target patient can be determined by the following formula 2: Formula 2 In formula 2, The first cardiovascular adverse presentation coefficient is used to characterize the target patient. Used to characterize the intensity of chest pain response in target patients. The norm is used to characterize the cardiac factor preference of the target patient, and it is used to characterize the normalization process, for example, the Softmax function.

[0070] Among them, the greater the intensity of chest pain response in the target patient in the short term and the lower the real-time cardiac factor dominance, the more significant the real-time adverse cardiovascular condition of the target patient.

[0071] Through the above technical solution, this invention refines the assessment of the target patient's real-time cardiovascular status into two independent dimensions: chest pain and heart rate. These are quantified separately by the chest pain assessment unit 111a and the heart rate assessment unit 111b to obtain the intensity of chest pain response and the preponderance of cardiac factors. This decomposed assessment method allows for a more comprehensive and refined capture of the patient's cardiovascular status. Subsequently, the cardiovascular assessment unit 111c fuses these two finely quantified indicators to determine a more accurate and robust first cardiovascular adverse presentation coefficient. Compared to directly using the original chest pain rating and real-time heart rate data, this solution can more effectively filter data noise, capture subtle changes in the patient's cardiovascular status, and significantly improve the assessment accuracy of the first cardiovascular adverse presentation coefficient.

[0072] In some of the embodiments of the present invention described above, relying solely on the patient's own reported first chest pain rating and chest pain frequency may not accurately and objectively measure the true intensity of the chest pain, especially in the absence of effective reference standards. This may lead to a bias in the assessment of the intensity of chest pain and affect the accuracy of the subsequent first cardiovascular adverse presentation coefficient.

[0073] In this regard, the present invention further proposes that the chest pain assessment unit 111a is used for: The mean of each first chest pain rating of the target patients was obtained by averaging the first chest pain ratings. The mean of the second chest pain ratings for each monitored patient was calculated. The chest pain rating score for the target patient is obtained by multiplying the ratio of the mean of the first chest pain rating to the mean of the second chest pain rating by the frequency of chest pain. The chest pain rating values ​​of the target patients were normalized to obtain the intensity of their chest pain response.

[0074] In this embodiment, after stent implantation, excessive proliferation of the vascular intima leads to narrowing of the stent lumen. Furthermore, as a foreign body, the stent stimulates platelet aggregation, which may form a thrombus that blocks the stent, thereby obstructing the transmission of blood to the myocardium. This obstructive mechanism causes patients to experience ischemic pain after the procedure. Therefore, if patients experience strong and frequent ischemic pain in a short period of time during the recovery period, it may reflect a poor cardiovascular condition and should be taken seriously.

[0075] To determine the intensity of chest pain response in target patients, firstly, the mean values ​​of all first chest pain ratings reported by target patients within a preset time period after stent implantation are averaged to obtain a mean first chest pain rating representing the average level of their chest pain condition. This averaging can be done using an arithmetic mean method, which involves summing all first chest pain rating values ​​and dividing by the number of ratings. Secondly, the mean values ​​of all second chest pain ratings reported by each monitored patient within the same preset time period are averaged to obtain a mean second chest pain rating representing the average level of chest pain condition in the monitored patient group. The monitored patients are typically a patient group with similar backgrounds, and their mean second chest pain ratings can serve as a benchmark for assessing the chest pain condition of the target patient. This averaging can also be done using an arithmetic mean method, averaging all second chest pain ratings in the monitored patient group. Finally, the ratio of the obtained mean first chest pain rating to the mean second chest pain rating is multiplied by the frequency of chest pain experienced by the target patient within the preset time period to obtain the target patient's chest pain evaluation value. This step compares the target patient's average chest pain rating with the average chest pain rating of monitored patients, incorporating the frequency of chest pain occurrence to obtain a comprehensive, relative index of chest pain severity. The ratio reflects the severity of the target patient's chest pain relative to the control group, while multiplying by frequency considers the density of chest pain occurrence. Finally, the target patient's chest pain rating is normalized (e.g., using a softmax function) to obtain the target patient's chest pain response intensity. Normalization aims to transform the chest pain rating to a uniform, comparable scale, eliminating the influence of dimensions, making it easier for subsequent calculations and comparisons, and ensuring that the chest pain response intensity remains within a specific range.

[0076] This invention introduces the mean of the second chest pain rating of monitored patients as a reference benchmark, making the chest pain assessment of the target patient no longer an isolated individual judgment, but possessing the objectivity of group comparison. First, the first chest pain rating of the target patient is averaged to smooth the volatility of individual reports and obtain a stable representation of their chest pain status. Simultaneously, the second chest pain rating of the monitored patient group is averaged to establish a representative baseline for chest pain under normal or stable conditions. Then, by calculating the ratio of the mean first chest pain rating of the target patient to the mean second chest pain rating of the monitored patients, the relative severity of the target patient's chest pain can be quantified, i.e., the degree to which their chest pain status deviates from the normal baseline. Multiplying this ratio by the frequency of chest pain in the target patient comprehensively considers the severity and density of chest pain, forming a more comprehensive chest pain evaluation value. Finally, this chest pain evaluation value is normalized, mapping it to a uniform scale to ensure the stability and comparability of chest pain response intensity in subsequent calculations.

[0077] As an example, the intensity of the target patient's chest pain response can be determined using the following formula 3: Formula 3 In formula 3, Used to characterize the intensity of chest pain response in target patients. Used to characterize the frequency of chest pain in target patients. The mean of the first chest pain rating used to characterize the target patients. The mean of the second chest pain rating is used to characterize the monitored patients, and norm is used to characterize the normalization process, for example, it can be the Softmax function.

[0078] Among them, the more frequent the chest pain and the higher the subjective pain level of the target patient in the short term, the higher the risk of adverse cardiovascular conditions for the target patient.

[0079] Through the aforementioned technical solution, the chest pain assessment unit, when determining the intensity of chest pain response in a target patient, no longer relies solely on the patient's own reported data. Instead, it incorporates the mean chest pain rating of the monitored patient population as a benchmark. This comparative assessment method significantly enhances the objectivity and accuracy of chest pain response intensity, effectively avoiding inaccurate assessments caused by individual subjective differences or reporting bias. Simultaneously, by combining the ratio of the mean chest pain rating with chest pain frequency, the severity and frequency of chest pain are comprehensively considered, allowing the chest pain response intensity to more comprehensively reflect the target patient's true chest pain burden. The final normalization process ensures good comparability of chest pain response intensity across different patients and time points, providing a more accurate and reliable input for the subsequent cardiovascular assessment unit to determine the first cardiovascular adverse event presentation coefficient. This improves the accuracy of the entire post-stent stent-related long-term adverse cardiovascular event risk dynamic prediction system in assessing the patient's immediate cardiovascular status, thereby enhancing the reliability of risk prediction.

[0080] In some embodiments of the present invention described above, a method is proposed to determine the cardiac factor prevalence of a target patient using a heart rate assessment unit 111b to characterize the degree of excellence of the heart rate status, and then to determine the first cardiovascular adverse presentation coefficient. However, in practical applications, simply relying on statistics or averages of real-time heart rate may not accurately capture abnormal fluctuations in heart rate and their potential impact on cardiovascular status, especially in long-term monitoring of patients after stent placement. Transient abnormalities or persistent deviations from the normal range in heart rate are often key early warning signals, requiring more refined assessment methods.

[0081] In this regard, the present invention further proposes that the heart rate assessment unit 111b is used for: Based on the target patient’s real-time heart rate within a preset time after stent implantation, extract abnormal heart rate segments where the real-time heart rate is lower than the first heart rate threshold or higher than the second heart rate threshold; the second heart rate threshold is greater than the first heart rate threshold. For each abnormal heart rate segment, the absolute value of the difference between the average heart rate of the first heart rate within the abnormal heart rate segment and the average heart rate of the second heart rate within a preset duration is taken to obtain the heart rate deviation value of each abnormal heart rate segment. The mean heart rate deviation values ​​of each abnormal heart rate segment were averaged to obtain the mean heart rate deviation. The priority of cardiac factors for the target patient is determined based on the number of abnormal segments of heart rate and the mean heart rate deviation.

[0082] In this embodiment, heart rate and vital signs data in actual post-stent implantation scenarios are important indicators reflecting adverse cardiovascular conditions. When the heart rate of a patient after stent implantation consistently exceeds 100 beats / minute, it increases myocardial oxygen consumption, aggravates the burden on the heart, and may induce myocardial ischemia. Conversely, if the heart rate consistently falls below 50 beats / minute, it may indicate insufficient cardiac output and decreased cardiac pumping function. Therefore, excessively fast or slow heart rate may exacerbate the burden on the heart or reflect an unfavorable cardiac condition. Thus, in order to improve the accuracy of the analysis while meeting the timeliness of risk assessment, it is necessary to combine the patient's heart rate and vital signs for auxiliary analysis.

[0083] Real-time heart rate within a preset duration refers to the heart rate data collected during continuous or periodic monitoring of the target patient's heart rate within a pre-defined time period after stent implantation. This preset duration can be flexibly set according to clinical needs or monitoring objectives, for example, it could be 24 hours. Real-time heart rate data can be obtained through various methods such as wearable devices, bedside monitors, or implantable devices.

[0084] Extracting anomalous heart rate segments where the real-time heart rate is below a first heart rate threshold or above a second heart rate threshold, with the second threshold being greater than the first, aims to identify abnormal periods where the target patient's heart rate deviates from the normal physiological range. The first and second heart rate thresholds represent the lower and upper limits of the heart rate, respectively, and can be set based on the patient's individual circumstances, age, underlying diseases, and clinical guidelines. For example, the first heart rate threshold could be set to 50 beats per minute, and the second heart rate threshold to 100 beats per minute. An anomalous heart rate segment refers to a period where the real-time heart rate is consistently below the first heart rate threshold or above the second heart rate threshold, or a point where the instantaneous heart rate exceeds these thresholds.

[0085] The first mean heart rate within each abnormal heart rate segment refers to the average value calculated by averaging all real-time heart rate data within that segment. This mean reflects the average level of abnormal heart rate within that segment. The second mean heart rate within a preset duration refers to the average value calculated by averaging all real-time heart rate data of the target patient over the entire preset duration. This mean represents the overall heart rate level of the target patient during that monitoring period and can serve as a benchmark for measuring the degree of heart rate deviation.

[0086] The absolute value of the difference is taken to obtain the heart rate deviation value. This operation is used to quantify the degree of abnormality of each abnormal heart rate segment. By comparing the first average heart rate within the abnormal heart rate segment with the second average heart rate over the entire preset duration and taking the absolute difference, the degree of heart rate deviation of that abnormal segment can be obtained.

[0087] The heart rate deviation values ​​of each abnormal heart rate segment are averaged to obtain the mean heart rate deviation. This step aims to comprehensively assess the average degree of deviation of all abnormal heart rate segments. By averaging the heart rate deviation values ​​of all identified abnormal heart rate segments, an indicator that better represents the overall degree of heart rate abnormality in the target patient can be obtained.

[0088] The number of anomalous heart rate segments refers to the total number of independent time intervals within a preset duration where the target patient's heart rate deviates from the normal range. This number reflects the frequency of heart rate abnormalities. Heart factor superiority is a comprehensive indicator used to characterize the excellence of the target patient's heart rate status. It is calculated by combining the number of anomalous heart rate segments and the mean heart rate deviation, providing a more comprehensive reflection of heart rate stability and the degree of abnormality.

[0089] When determining the cardiac factor eugenics of a target patient, the aforementioned heart rate assessment unit 111b first identifies and extracts all abnormal heart rate segments that deviate from the normal range based on the target patient's real-time heart rate data within a preset time period after stent implantation, by setting a first heart rate threshold and a second heart rate threshold. This method avoids the possibility of instantaneous or periodic abnormalities being masked by relying solely on the overall average heart rate. Subsequently, for each identified abnormal heart rate segment, its internal average heart rate (first average heart rate) is calculated and compared with the average heart rate over the entire preset time period (second average heart rate). The degree of deviation of each abnormal segment is quantified by taking the absolute difference, thereby obtaining a heart rate deviation value. This comparison method can objectively reflect the severity of heart rate abnormalities. Next, the heart rate deviation values ​​of all abnormal heart rate segments are averaged to obtain the average heart rate deviation value, in order to comprehensively assess the average level of heart rate abnormality. At the same time, the number of abnormal heart rate segments is counted to reflect the frequency of heart rate abnormalities. Ultimately, the heart rate assessment unit comprehensively utilizes the number of abnormal heart rate segments and the mean heart rate deviation to determine the cardiac factor prevalence of the target patient. This comprehensive assessment mechanism ensures that cardiac factor prevalence considers not only the severity of heart rate abnormalities but also the frequency of their occurrence, thus providing a more comprehensive and precise characterization of the target patient's cardiac rate status.

[0090] As an example, the priority of cardiac factors for the target patient can be determined using the following formula 4: Formula 4 In formula 4, The value of cardiac factors is used to characterize the prevalence of the target patient, and M is used to characterize the number of abnormal segments in the target patient's heart rate. Used to characterize the mean heart rate deviation of the target patient.

[0091] Among them, the fewer abnormal heart rate segments appearing in the target patient after surgery, and the lower the numerical deviation of heart rate, the better the real-time cardiovascular status of the target patient is reflected from the heart rate perspective.

[0092] It should be noted that when the number of abnormal segments extracted from the heart rate abnormal segment is 0, it means that no heart rate abnormality has occurred. In this case, there is no need to calculate, and the heart factor priority is directly assigned to the preset maximum constant (e.g., =1000).

[0093] Through the above technical solution, this invention enables a more refined assessment of the heart rate status of patients after stent placement. By identifying and quantifying the frequency and severity of anomalous heart rate segments, it avoids potential risk signals that may be masked by simple heart rate averaging in traditional methods. This in-depth analysis of heart rate abnormalities allows cardiac factor dominance to more accurately reflect the immediate changes and long-term trends in the patient's cardiovascular status, thus providing a more reliable basis for determining the first adverse cardiovascular event presentation coefficient. This significantly improves the early identification capability and prediction accuracy of the dynamic prediction system for long-term adverse cardiovascular events after stent placement, facilitating timely intervention and improving patient prognosis.

[0094] In some embodiments of the present invention described above, the system uses the cardiac function assessment module 120 to determine the emphasis of cardiac function analysis at each stage of the target patient's post-stent implantation physical examination data. However, in practical applications, if the types of physical examination data are not specific enough or the processing methods are not refined enough, it may lead to an incomplete and inaccurate assessment of the patient's overall cardiac function, especially in identifying the risk of thrombosis closely related to post-stent implantation, thereby affecting the accuracy of long-term adverse cardiovascular event risk prediction.

[0095] In this regard, such as Figure 4 As shown, the present invention further proposes that the physical examination data include the mean ST segment value of electrocardiogram, platelet count, left ventricular ejection fraction and coronary angiography stenosis degree; Cardiac function assessment module 120 includes: Thrombosis assessment submodule 121 is used to determine the thrombosis trend of the target patient at the target stage based on the mean ST segment value and platelet count of the target patient's electrocardiogram at the target stage after stent implantation. The cardiac function assessment submodule 122 is used to determine the emphasis of cardiac function analysis in the target patient at the target stage based on thrombosis trend, left ventricular ejection fraction, and coronary angiography stenosis.

[0096] In this embodiment, the mean ST segment value on electrocardiogram (ECG) is an important indicator reflecting myocardial ischemia or damage, and its changes may indicate the risk of myocardial infarction or restenosis. This mean value can be obtained by statistically averaging multiple ECG measurements of a patient over a specific time period, or by automatically collecting and calculating it using a continuous electrocardiogram (ECG) monitoring device.

[0097] Platelet count is the number of platelets in the blood, which is closely related to blood clotting and thrombosis. Both excessively high and low counts can increase the risk of cardiovascular events. This count is usually obtained through routine blood tests, but can also be monitored in real time using automated blood analyzers.

[0098] Left ventricular ejection fraction (LVEF) is a core indicator of cardiac pumping function, reflecting the proportion of blood pumped out by the left ventricle with each contraction. It is a key parameter for assessing the degree of heart failure. This score can be obtained through imaging examinations such as echocardiography, cardiac magnetic resonance imaging (MRI), or computed tomography (CT).

[0099] Coronary angiography shows that the degree of stenosis directly reflects the severity of coronary atherosclerosis and is an important basis for assessing the progression and prognostic risk of coronary heart disease. This degree of stenosis is usually quantitatively assessed through coronary angiography or coronary CT angiography.

[0100] The thrombosis assessment submodule 121 is a functional unit within the cardiac function assessment module 120. Its main function is to specifically assess the potential risk of thrombosis. This thrombosis assessment submodule 121 receives the mean ST segment value from the electrocardiogram and platelet count as input. Through a preset algorithm model, it comprehensively analyzes the changing trends and interrelationships of these two indicators, thereby outputting a quantified thrombosis trend degree.

[0101] The cardiac function assessment submodule 122 is another core functional unit of the cardiac function assessment module 120. Its responsibility is to integrate thrombosis trend and other key cardiac function indicators, such as left ventricular ejection fraction and coronary angiography stenosis, to generate a comprehensive cardiac function analysis that reflects the overall condition of the heart.

[0102] The present invention specifies physical examination data as mean ST segment on electrocardiogram (ECG), platelet count, left ventricular ejection fraction (LVEF), and coronary angiography stenosis. It introduces a thrombosis assessment submodule 121 and a cardiac function assessment submodule 122, enabling the cardiac function assessment module 120 to more precisely capture changes in cardiac function in target patients after stent implantation. The thrombosis assessment submodule 121 specifically analyzes the mean ST segment on ECG and platelet count, effectively identifying potential risks of thrombosis, which is particularly crucial for post-stent patients. Subsequently, the cardiac function assessment submodule 122 integrates the thrombosis trend with comprehensive indicators such as LVEF and coronary angiography stenosis, generating a more comprehensive and accurate cardiac function analysis focus. This hierarchical and refined assessment mechanism ensures that the overall assessment of the patient's cardiac function not only covers routine cardiac function indicators but also pays special attention to the risks of post-stent complications, allowing the cardiac function analysis focus to more accurately reflect the patient's immediate and long-term cardiovascular health status.

[0103] As an example, the emphasis of cardiac function analysis in the target patient at the target stage can be determined using the following formula 5: Formula 5 In formula 5, G is used to characterize the emphasis of cardiac function analysis in the target patient at the target stage, and G is used to characterize the thrombosis trend in the target patient at the target stage. The parameter is used to characterize the coronary angiography stenosis of the target patient at the target stage, H is used to characterize the left ventricular ejection fraction of the target patient at the target stage, and norm is used to characterize the normalization process, for example, it can be the Softmax function.

[0104] The greater the tendency of thrombosis in the target patient, the greater the emphasis on their cardiac function analysis. Furthermore, poor cardiovascular condition leads to acute vascular occlusion, which in turn affects ventricular systolic function, reflected in a decreased left ventricular ejection fraction and increased coronary angiographic stenosis. Therefore, the smaller the left ventricular ejection fraction and the greater the coronary angiographic stenosis, the greater the emphasis on their cardiac function analysis.

[0105] Through the aforementioned technical solutions, the cardiac function assessment module can utilize more specific and clinically significant physical examination data and employ a stratified assessment mechanism to conduct a more comprehensive, detailed, and accurate evaluation of the cardiac function of target patients after stent implantation. This not only improves the reliability of the emphasis in cardiac function analysis, enabling it to more realistically reflect the overall status of the patient's cardiac function, but also, through specialized thrombosis assessment, can identify the potential risk of thrombosis after stent implantation earlier and more effectively. This provides a more solid data foundation for determining the probability of the first adverse event, significantly improving the overall predictive accuracy and clinical practical value of the dynamic prediction system for long-term adverse cardiovascular event risk after stent implantation.

[0106] In some embodiments of the present invention described above, a cardiac function assessment module 120 is proposed to determine the emphasis of cardiac function analysis at each stage based on the physical examination data of the target patient at each stage after stent implantation. The physical examination data includes the mean ST segment value of the electrocardiogram and platelet count. A thrombosis trend is determined through a thrombosis assessment submodule 121. However, in practical applications, using only the raw values ​​of the mean ST segment value of the electrocardiogram and platelet count to assess the thrombosis trend may not adequately account for individual differences and the relative significance of these indicators in different patient groups, thus affecting the accuracy and sensitivity of thrombosis risk assessment.

[0107] In this regard, the present invention further proposes that the thrombosis assessment submodule 121 is used for: The absolute value of the difference between the mean ST segment value of the target patient's ECG at the target stage and the maximum ST segment value of the ECG in each monitored patient is divided by the maximum ST segment value of the ECG to obtain the ECG ST segment variability. The platelet count variability is obtained by dividing the absolute value of the difference between the platelet count of the target patient at the target stage and the maximum platelet count among all monitored patients by the maximum platelet count. Based on the ST segment variability of electrocardiogram and the platelet count variability, the thrombosis trend of the target patient at the target stage is determined.

[0108] In this embodiment, the maximum ST segment value on the electrocardiogram (ECG) of monitored patients refers to the maximum mean ST segment value observed from a group of monitored patients with similar clinical characteristics within the same target stage, serving as a reference benchmark for assessing the severity of myocardial ischemia. This maximum value can be extracted from a historical medical database by screening out eligible monitored patient groups.

[0109] The ST segment variability of electrocardiogram is a dimensionless index obtained by dividing the absolute value of the difference between the mean ST segment value of the target patient's electrocardiogram and the maximum ST segment value of the monitored patients' electrocardiograms by the maximum ST segment value of the electrocardiogram. It quantifies the relative severity of myocardial ischemia in the target patient.

[0110] The maximum platelet count in monitored patients refers to the highest platelet count observed within a group of monitored patients with similar clinical characteristics during the same target phase, serving as a reference benchmark for measuring platelet abnormalities. This maximum value can be filtered and extracted from historical medical databases.

[0111] Platelet count variability is a dimensionless index obtained by dividing the absolute value of the difference between the platelet count of the target patient and the maximum platelet count in the monitored patients by the maximum platelet count. It quantifies the relative degree of abnormality in the platelet level of the target patient.

[0112] Thrombosis trend score is an indicator determined by combining ST segment variability on electrocardiogram (ECG) and platelet count variability. It is used to characterize the tendency or risk of thrombosis in a target patient at a specific target stage. This trend score can be calculated by fusing ECG ST segment variability and platelet count variability using various algorithms such as weighted averaging, fuzzy logic reasoning, and machine learning models; or it can be combined and scored according to a pre-set risk scoring table.

[0113] The present invention introduces the concept of "variability" and uses the maximum value in the monitored patient population as a reference, enabling the thrombosis assessment submodule 121 to more precisely capture the relative deviation of the target patient's ECG ST segment and platelet count from the high-risk group when determining the thrombosis trend. Specifically, the calculation of ECG ST segment variability, by comparing and normalizing the mean ECG ST segment value of the target patient with the maximum ECG ST segment value of the monitored patients, effectively quantifies the relative severity of myocardial ischemia in the target patient. Similarly, the calculation of platelet count variability, by comparing and normalizing the platelet count value with the maximum platelet count of the monitored patients, reflects the relative degree of abnormality in the target patient's platelet level. These two variability indicators, compared to the original values, better reflect the target patient's current relative position and risk level within the group. Subsequently, the thrombosis assessment submodule 121 comprehensively determines the thrombosis trend based on these two relatively meaningful variability indicators. This approach allows the assessment of thrombosis trend to no longer rely solely on absolute values, but incorporates the concepts of population reference and relative change, thereby improving the accuracy and objectivity of the assessment.

[0114] As an example, the thrombosis tendency of the target patient at the target stage can be determined using the following formula 6: Formula 6 In formula 6, Used to characterize the thrombosis trend in target patients at the target stage. Used to characterize the maximum ST segment value on the electrocardiogram of each monitored patient. Used to characterize the mean ST segment value of the electrocardiogram of the target patient at the target stage. Used to characterize the maximum platelet count in each monitored patient. Used to characterize the platelet count of the target patient at the target stage. and For the normalization preset weights, all are selected as 0.5 in this invention, and can be adjusted according to the actual scenario; exp is used to characterize the exponential function operation.

[0115] Among them, the more significant the abnormal electrocardiogram and platelet count of the target patient in the next stage of physical examination, the greater the tendency to thrombosis.

[0116] Through the above technical solution, the thrombosis assessment submodule 121, when assessing the thrombosis trend, no longer relies solely on the absolute values ​​of the mean ST segment on the electrocardiogram and the platelet count. Instead, it introduces the concept of "variability" that is compared with and normalized to the maximum values ​​of the monitored patient population. This relativistic assessment method can effectively eliminate the influence of individual differences on the assessment results, enabling the thrombosis trend to more accurately reflect the relative risk level of the target patient within the population.

[0117] In some of the embodiments of the present invention described above, simply relying on the target cardiovascular adverse presentation coefficient and the emphasis of cardiac function analysis may not be sufficient to capture the dynamic changes in the patient's cardiovascular status. In particular, when assessing long-term risk, the lack of consideration for the evolution of cardiac function over time may lead to insufficient accuracy and sensitivity in risk prediction.

[0118] In response, the present invention further proposes that the probability determination module 130 is used for: The emphasis of each cardiac function analysis was fitted with a straight line in chronological order to obtain the target fitted straight line; Based on the slope of the target fitted line, the target cardiac function analysis emphasis, and the target cardiovascular adverse event presentation coefficient, the probability of the first adverse event for the target patient is determined; the target cardiac function analysis emphasis is the cardiac function analysis emphasis closest to the current time among all cardiac function analysis emphasis.

[0119] In this embodiment, the emphasis of each cardiac function analysis is linearly fitted in chronological order to obtain a target fitted line. This step aims to reveal the long-term changing patterns of cardiac function by performing trend analysis on the patient's cardiac function analysis emphasis data at different time points. The emphasis of cardiac function analysis is a key indicator characterizing the overall status of a patient's cardiac function, and its trend over time is significant for assessing long-term risk. In practice, the least squares method can be used to perform linear regression on a series of time-ordered cardiac function analysis emphasis data points to obtain a straight line that best reflects the changing trends of these data points, i.e., the target fitted line.

[0120] The probability of the first adverse event for the target patient is determined by multiplying the slope of the target fitted line, the emphasis of the target cardiac function analysis, and the target cardiovascular adverse event presentation coefficient. This step aims to comprehensively consider the long-term trend of the patient's cardiac function (reflected by the slope of the line), the current cardiac function status (reflected by the emphasis of the target cardiac function analysis), and the immediate cardiovascular status (reflected by the target cardiovascular adverse event presentation coefficient) to calculate a comprehensive risk probability.

[0121] The target cardiac function analysis emphasis is the cardiac function analysis emphasis most recent to the current time among all cardiac function analysis emphasis values. This feature clarifies the specific selection principle of the "target cardiac function analysis emphasis" used when calculating the probability of the first adverse event, namely, selecting the most recently measured cardiac function analysis emphasis. This ensures that the risk assessment can reflect the patient's current cardiac function status in a timely manner, improving the real-time performance and accuracy of the prediction. The system can maintain a list of cardiac function analysis emphasis values ​​stored by timestamps. When calculation is needed, the system can directly retrieve and select the cardiac function analysis emphasis with the timestamp closest to the current time from the list.

[0122] The present invention, through the probability determination module 130, considers not only the target cardiovascular adverse event presentation coefficient and the emphasis of each cardiac function analysis when determining the probability of the first adverse event in a target patient. Furthermore, it obtains a target fitted line by linearly fitting the emphasis of each cardiac function analysis in chronological order. This target fitted line intuitively reflects the trend of the patient's cardiac function changes over time. Subsequently, based on the slope of this target fitted line, the most recent target cardiac function analysis emphasis, and the target cardiovascular adverse event presentation coefficient, the system finally obtains the probability of the first adverse event in the target patient. The ingenuity of this approach lies in its organic combination of the patient's immediate cardiovascular state (target cardiovascular adverse event presentation coefficient), the latest cardiac function state (target cardiac function analysis emphasis), and the long-term evolution trend of cardiac function (slope of the line). The slope of the line quantifies the rate of deterioration or improvement of cardiac function, the target cardiac function analysis emphasis provides the baseline state at the current moment, and the target cardiovascular adverse event presentation coefficient reflects the immediate risk of cardiovascular events. By taking these three key dimensions into account, the system can generate a more comprehensive, dynamic and forward-looking probability of the first adverse event, thereby overcoming the limitations of prediction based solely on static or discrete data, and making the assessment of patients' long-term risks more accurate and sensitive.

[0123] As an example, the probability of the first adverse event in the target patient can be determined using the following formula 7: Formula 7 In formula 7, Used to characterize the probability of the first adverse event in the target patient. The slope of the line used to characterize the target fitted line. Used to characterize the emphasis of target cardiac function analysis The coefficient is used to characterize the presentation of the target cardiovascular adverse events. `norm` is used to characterize the normalization process, for example, it could be the Softmax function. `exp` is used to characterize the exponential function processing.

[0124] Among them, the higher the real-time target cardiovascular adverse event presentation coefficient of the current target patient, and the more significant the deterioration of cardiovascular status reflected by multi-stage physical examination, the greater the real-time probability of cardiovascular adverse events in the target patient.

[0125] It should be noted that if the current target patient has had fewer than two historical physical examinations, linear fitting cannot be performed. In this case, the slope K of the line should be set to the preset baseline value of 1, since linear fitting requires at least two data points.

[0126] By employing the aforementioned technical solution, when determining the probability of the first adverse event in a target patient, the system not only considers the patient's immediate cardiovascular status and current cardiac function, but also incorporates trend information on cardiac function changes over time. This trend analysis enables the predictive model to capture potential deterioration or improvement in cardiac function, thereby enhancing the dynamism and foresight of the first adverse event probability. By comprehensively considering immediate status, latest status, and long-term trends, the system can more accurately and sensitively assess the risk of long-term adverse cardiovascular events after stent placement, providing a more reliable basis for subsequent risk warning and intervention, and effectively avoiding the problems of predictive lag or inaccuracy caused by relying solely on static data.

[0127] In some of the embodiments of the present invention described above, simply comparing probabilities may not provide an intuitive, standardized and comparable risk warning indicator. In particular, when the number of monitored patients is large or the probability distribution of their second adverse events is uneven, direct comparison is difficult to accurately reflect the true degree of deviation of the target patient from the group risk, thereby affecting the effectiveness and accuracy of risk warning.

[0128] In response, the present invention further proposes that the index determination module 140 is used for: The mean of the second adverse event probability for each monitored patient was calculated. The difference between the probability of the first adverse event and the mean probability of adverse events in the target patients is normalized to obtain the target cardiovascular risk warning index for the target patients.

[0129] In this embodiment, mean processing is a statistical method designed to obtain the central tendency of a set of values ​​by calculating the average of the values. In this technical solution, mean processing is applied to the second adverse event probabilities of each monitored patient to comprehensively consider the risk levels of multiple monitored patients, thereby establishing a representative population risk benchmark. This mean processing can employ an arithmetic mean method, which involves summing the second adverse event probabilities of all monitored patients and dividing by the total number of monitored patients. The mean of adverse event probabilities obtained through mean processing provides a stable reference point for subsequent risk assessment of target patients. The difference is used to quantify the deviation between the first adverse event probability of the target patient and the mean of adverse event probabilities, thereby objectively reflecting the anomaly of the target patient's risk. Normalization is a data transformation technique designed to map the original data to a preset specific range (e.g., 0 to 1 or 0 to 100) to eliminate the influence of different data units and ranges, making the data comparable. For example, the difference can be input into the Sigmoid mapping function to perform interval normalization from 0 to 1. The value obtained after normalization is the target cardiovascular risk warning indicator. This indicator is a standardized, easy-to-understand and compare quantitative value that can intuitively reflect the cardiovascular risk level of the target patient.

[0130] It should be noted that the calculation methods for the first adverse event probability and the second adverse event probability are the same, the only difference being the corresponding objects. The first adverse event probability is the adverse event probability for the target patient, and the second adverse event probability is the adverse event probability for the monitored patient.

[0131] This invention establishes a statistically significant population risk benchmark—the mean adverse event probability—by introducing a mean processing method for the probability of a second adverse event in monitored patients. This benchmark effectively integrates risk information from multiple monitored patients, avoiding the randomness and instability that may arise from monitoring a single patient or making simple comparisons, thus providing a more robust reference system for assessing the risk of target patients. Based on this, by calculating the difference between the probability of a first adverse event in the target patient and the mean adverse event probability, the deviation of the individual risk of the target patient from the population average risk can be accurately quantified. This quantification of deviation allows the system to identify anomalies in the risk of the target patient. Subsequently, this difference is normalized, transforming the original risk deviation value into a standardized, dimensionless indicator—the target cardiovascular risk warning indicator. This normalization ensures good comparability and interpretability of risk assessment results across different patients, time points, or monitored populations, making risk warnings more intuitive and easier to understand.

[0132] Through the above technical solution, this invention overcomes the limitations of simple probability comparisons, providing a more scientific, objective, and comparable cardiovascular risk early warning indicator. Specifically, by averaging the probability of second adverse events in monitored patients, the system establishes a stable population risk benchmark, effectively reducing the randomness of assessment. Based on this, by calculating and normalizing the difference between the target patient and this benchmark, the risk level of the target patient can be accurately quantified and standardized, thus intuitively reflecting the degree of risk deviation relative to the same population. This allows medical staff to more clearly understand the risk status of the target patient and conduct dynamic monitoring and intervention based on standardized indicators, significantly improving the accuracy and practicality of predicting long-term adverse cardiovascular event risks after stent placement.

[0133] In some of the embodiments of the present invention described above, obtaining only the target cardiovascular risk warning indicators cannot directly provide patients or medical staff with clear and actionable risk management advice, which may lead to misjudgment of risks or untimely response, thereby affecting the patient's long-term prognosis.

[0134] In response, this invention further proposes that, after comparing the probability of the first adverse event in the target patient with the probability of the second adverse event in each monitored patient to obtain the target cardiovascular risk warning index for the target patient, the invention also includes: The indicator analysis module is used to compare the target cardiovascular risk warning indicators of the target patient with the first warning indicator threshold and the second warning indicator threshold to obtain the indicator comparison results; the second warning indicator threshold is greater than the first warning indicator threshold. The indicator analysis module is also used to initiate a first warning reminder to the target patient's communication terminal in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is greater than the second warning indicator threshold; the first warning reminder is the highest level of warning reminder; The indicator analysis module is also used to continue risk prediction for the target patient in response to the indicator comparison results indicating that the target cardiovascular risk warning indicator is less than the first warning indicator threshold. The indicator analysis module is also used to initiate a second warning reminder corresponding to the target cardiovascular risk warning indicator to the target patient's communication terminal in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is not less than the first warning indicator threshold and not greater than the second warning indicator threshold.

[0135] In this embodiment, the indicator analysis module is a key component of the system used for intelligent interpretation and decision-making regarding the calculated target cardiovascular risk warning indicators. Its core function is to transform the target cardiovascular risk warning indicators into risk levels with clear action guidance significance.

[0136] The first and second warning indicator thresholds are critical values ​​used to classify risk levels. These thresholds can be derived from statistical analysis of extensive clinical data, such as by training machine learning models on historical patient data to identify the range of indicators corresponding to different risk levels; or they can be set by medical experts based on clinical experience and guidelines. The first warning indicator threshold typically represents a lower risk threshold, while the second warning indicator threshold represents a higher risk threshold; together, they construct a tiered risk assessment system. For example, the first warning indicator threshold could be 0.58, and the second warning indicator threshold could be 0.85.

[0137] The indicator comparison result is a judgment made by the indicator analysis module after comparing the target cardiovascular risk warning indicator with a preset threshold. This result can be a Boolean combination indicating whether the indicator exceeds a certain threshold. This result serves as the basis for subsequent implementation of different warning measures.

[0138] In response to a comparison result indicating that the target cardiovascular risk warning indicator exceeds the second warning indicator threshold, a first warning alert is initiated to the target patient's communication terminal. This means that when the risk reaches the highest level, the system will immediately trigger the most urgent notification mechanism. This communication terminal can be the patient's smartphone, smart wearable device, or home gateway device. As the highest level of warning alert, the first warning alert can be sent via multiple channels simultaneously, including but not limited to: SMS, app push notifications, and telephone voice notifications, possibly accompanied by a loud alarm sound, or even directly notifying the patient's designated emergency contact or medical staff to ensure the patient receives timely medical intervention.

[0139] In response to the indicator comparison results indicating that the target cardiovascular risk warning indicator is less than the first warning indicator threshold, the system continues to predict the risk of the target patient. This means that when the risk is at a low level, the system determines that no additional intervention is needed and only the routine risk monitoring process needs to be maintained. This means that the system will continue to collect data and calculate risk indicators periodically, but will not trigger any warning notifications, thus avoiding unnecessary disruption.

[0140] In response to the indicator comparison results indicating that the target cardiovascular risk warning indicator is not less than the first warning indicator threshold and not greater than the second warning indicator threshold, a second warning reminder corresponding to the target cardiovascular risk warning indicator is initiated to the target patient's communication terminal. This indicates that when the risk is at a moderate level, the system will issue a non-urgent but suggestive warning. This second warning reminder can provide personalized suggestions based on specific risk indicator values, such as reminding patients to pay attention to their lifestyle habits, suggesting regular check-ups or consulting a doctor, through app messages, emails, etc., but it is usually not as mandatory or urgent as the first warning reminder.

[0141] This invention introduces an indicator analysis module and a tiered early warning mechanism, enabling the system to intelligently assess and tier the risk based on preset thresholds after calculating the target cardiovascular risk early warning indicator. Once the target cardiovascular risk early warning indicator is calculated, the indicator analysis module immediately compares it with a first early warning threshold and a second early warning threshold. If the indicator value exceeds the second early warning threshold, indicating an extremely high risk of cardiovascular events, the system immediately triggers the highest-level first early warning alert to ensure the patient receives timely medical intervention. If the indicator value is below the first early warning threshold, the risk is considered low, and the system continues with routine risk prediction to avoid unnecessary alarms. When the indicator value falls between the two thresholds, the system issues a second early warning alert corresponding to the specific risk level, prompting the patient to pay attention and take appropriate preventative measures. This tiered response mechanism allows the system to provide differentiated and targeted feedback based on the severity of the risk, effectively guiding patients in self-management or seeking medical help.

[0142] Through the above technical solution, this invention can transform abstract risk warning indicators into specific and actionable risk management actions, achieving refined and intelligent management of patients' cardiovascular risks. This tiered warning mechanism avoids a "one-size-fits-all" approach, ensuring timely and urgent intervention for high-risk patients while preventing unnecessary panic and resource waste for low-risk patients, and providing targeted health guidance for moderate-risk patients. This significantly improves the practicality and effectiveness of the long-term adverse cardiovascular event risk prediction system for post-stent patients, helping to improve long-term patient prognosis and reduce the incidence of adverse events.

[0143] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0144] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system, characterized in that, The system includes: The cardiovascular assessment module is used to determine the target cardiovascular adverse presentation coefficient of the target patient based on the first chest pain rating and real-time heart rate after stent implantation; the target cardiovascular adverse presentation coefficient is used to characterize the immediate cardiovascular status of the target patient. The cardiac function assessment module is used to determine the emphasis of cardiac function analysis at each stage of the target patient's physical examination data after stent implantation; the emphasis of cardiac function analysis is used to characterize the overall status of the target patient's cardiac function. The probability determination module is used to determine the probability of the first adverse event in the target patient based on the target cardiovascular adverse event presentation coefficient and the emphasis of each of the cardiac function analyses; The indicator determination module is used to compare the probability of the first adverse event of the target patient with the probability of the second adverse event of each monitored patient to obtain the target cardiovascular risk warning indicator of the target patient.

2. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 1, wherein, The cardiovascular assessment module includes: The cardiovascular assessment submodule is used to determine the first cardiovascular adverse presentation coefficient of the target patient based on the first chest pain rating and real-time heart rate of the target patient after stent implantation; the first cardiovascular adverse presentation coefficient is the cardiovascular adverse presentation coefficient of the target patient after stent implantation. The deviation assessment submodule is used to subtract the baseline value of the cardiovascular adverse presentation coefficient from the first cardiovascular adverse presentation coefficient of the target patient to obtain the deviation value of the cardiovascular adverse presentation coefficient; the baseline value of the cardiovascular adverse presentation coefficient is the average value of the second cardiovascular adverse presentation coefficient of the target patient multiple times before stent implantation; the second cardiovascular adverse presentation coefficient is the cardiovascular adverse presentation coefficient of the target patient before stent implantation. The evaluation correction submodule is used to correct the first cardiovascular adverse presentation coefficient based on the deviation value of the cardiovascular adverse presentation coefficient, so as to obtain the target cardiovascular adverse presentation coefficient of the target patient.

3. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 2, wherein, The cardiovascular assessment submodule includes: A chest pain assessment unit is used to determine the intensity of the chest pain response of the target patient based on the frequency of chest pain within a preset time after stent implantation and the corresponding first chest pain rating. A heart rate assessment unit is used to determine the cardiac factor prevalence of the target patient based on the real-time heart rate of the target patient within a preset time after stent implantation; the cardiac factor prevalence is used to characterize the degree of excellence of the heart rate status. A cardiovascular assessment unit is used to determine the first cardiovascular adverse presentation coefficient of the target patient based on the intensity of the target patient's chest pain response and the priority of the cardiac factors.

4. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 3, wherein, The chest pain assessment unit is used for: The mean of each of the first chest pain ratings of the target patients is obtained by averaging the first chest pain ratings. The mean value of the second chest pain rating of each monitored patient was obtained by averaging the second chest pain rating. The chest pain evaluation score of the target patient is obtained by multiplying the ratio of the mean of the first chest pain rating to the mean of the second chest pain rating by the frequency of chest pain. The chest pain evaluation values ​​of the target patients were normalized to obtain the intensity of the chest pain response of the target patients.

5. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 3, wherein, The heart rate assessment unit is used for: Based on the real-time heart rate of the target patient within a preset time after stent implantation, extract the abnormal heart rate segments where the real-time heart rate is lower than a first heart rate threshold or higher than a second heart rate threshold. The second heart rate threshold is greater than the first heart rate threshold; For each abnormal heart rate segment, the absolute value of the difference between the average first heart rate within the abnormal heart rate segment and the average second heart rate within the preset duration is taken to obtain the heart rate deviation value for each abnormal heart rate segment. The mean heart rate deviation values ​​of each of the abnormal heart rate segments are averaged to obtain the mean heart rate deviation. The cardiac factor preference of the target patient is determined based on the number of abnormal heart rate segments and the mean heart rate deviation.

6. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 1, wherein, The physical examination data includes the mean ST segment value of electrocardiogram, platelet count, left ventricular ejection fraction, and coronary angiography stenosis. The cardiac function assessment module includes: The thrombosis assessment submodule is used to determine the thrombosis trend of the target patient at the target stage based on the mean ST segment value of the electrocardiogram and the platelet count of the target patient at the target stage after stent implantation. The cardiac function assessment submodule is used to determine the emphasis of cardiac function analysis in the target patient at the target stage based on the thrombosis trend, the left ventricular ejection fraction, and the coronary angiography stenosis.

7. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 6, wherein, The thrombosis assessment submodule is used for: The absolute value of the difference between the mean ST segment value of the electrocardiogram of the target patient in the target stage and the maximum ST segment value of the electrocardiogram of each monitored patient is divided by the maximum ST segment value of the electrocardiogram to obtain the ST segment variability of the electrocardiogram. The platelet count variability is obtained by dividing the absolute value of the difference between the platelet count of the target patient at the target stage and the maximum platelet count among the monitored patients by the maximum platelet count. Based on the ST segment variability of the electrocardiogram and the platelet count variability, the thrombosis trend of the target patient in the target stage is determined.

8. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 1, wherein, The probability determination module is used for: The emphasis of each cardiac function analysis is fitted with a straight line in chronological order to obtain the target fitted straight line; Based on the slope of the target fitted line, the emphasis of the target cardiac function analysis, and the target cardiovascular adverse event presentation coefficient, the probability of the first adverse event for the target patient is determined. The target cardiac function analysis emphasis is the cardiac function analysis emphasis that is closest to the current time among all the cardiac function analysis emphasis.

9. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 1, wherein, The indicator determination module is used for: The average probability of the second adverse event for each of the monitored patients was calculated to obtain the mean probability of the adverse event. The difference between the probability of the first adverse event and the mean probability of the adverse event in the target patient is normalized to obtain the target cardiovascular risk warning index for the target patient.

10. The stent post-procedure long-term adverse cardiovascular event risk dynamic prediction system of claim 1, wherein, After comparing the probability of a first adverse event in the target patient with the probability of a second adverse event in each monitored patient to obtain the target cardiovascular risk warning index for the target patient, the method further includes: The indicator analysis module is used to compare the target cardiovascular risk warning indicator of the target patient with the first warning indicator threshold and the second warning indicator threshold to obtain the indicator comparison result; the second warning indicator threshold is greater than the first warning indicator threshold. The indicator analysis module is also used to initiate a first warning reminder to the target patient's communication terminal in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is greater than the second warning indicator threshold; the first warning reminder is the highest level warning reminder; The indicator analysis module is also used to continue risk prediction for the target patient in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is less than the first warning indicator threshold; The indicator analysis module is also used to initiate a second warning reminder corresponding to the target cardiovascular risk warning indicator to the communication terminal of the target patient in response to the indicator comparison result indicating that the target cardiovascular risk warning indicator is not less than the first warning indicator threshold and not greater than the second warning indicator threshold.

Images