Coronary heart disease recurrence risk assessment method and device, equipment and storage medium

Abstract

The invention provides a coronary heart disease recurrence risk assessment method and device, equipment and a storage medium, and relates to the technical field of medical data processing. The method comprises the following steps: acquiring dynamic behavior data, physiological data and static risk indexes of a target patient; calculating a treatment compliance index and a rehabilitation health index of the target patient based on the dynamic behavior data and the physiological data of the target patient; and inputting the treatment compliance index, the rehabilitation health index and the static risk index into a trained coronary heart disease recurrence risk scoring model to obtain a coronary heart disease recurrence risk score of the target patient. According to the method, objective and quantitative recurrence risk scores can be obtained, more accurate decision support is provided for clinicians, and early warning and personalized intervention can be realized, so that the prognosis of patients is improved, and the medical cost is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical data processing, and in particular to a coronary heart disease recurrence risk assessment method, device, equipment and storage medium. BACKGROUND

[0002] Coronary heart disease (CHD) is one of the major diseases leading to death and disability worldwide, and its high recurrence rate is a serious challenge for clinical diagnosis and treatment. At present, the recurrence risk assessment of coronary heart disease mainly relies on static medical examination indicators such as low-density lipoprotein cholesterol, systolic blood pressure, and coronary CT stenosis score. These indicators can only reflect the physiological state at a single time point, and there is significant evaluation lag. And the existing evaluation method completely ignores the patient's compliance and dynamic rehabilitation process and other behavioral factors, and these factors are exactly the key determinants that affect the long-term prognosis of coronary heart disease. Due to the lack of systematic collection and integrated analysis of multi-dimensional dynamic data, the risk assessment dimension is single, the result is one-sided, and the patient's true risk state cannot be accurately reflected. In addition, the existing evaluation model uses a unified weight standard and does not consider the differences in individual characteristics such as age and medical history of different patients, further limiting the accuracy of the evaluation. This one-sided and static evaluation mode makes it difficult to provide effective early warning for the clinic, hindering the implementation of precise prevention and control strategies. SUMMARY

[0003] The embodiments of the present application provide a coronary heart disease recurrence risk assessment method, device, equipment and storage medium to solve the problem that it is difficult to accurately assess the recurrence risk of coronary heart disease at present.

[0004] In a first aspect, the embodiments of the present application provide a coronary heart disease recurrence risk assessment method, comprising: obtaining dynamic behavior data, physiological data and static risk indicators of a target patient; calculating a treatment compliance indicator and a rehabilitation health index of the target patient based on the dynamic behavior data and the physiological data of the target patient; inputting the treatment compliance indicator, the rehabilitation health index and the static risk indicator into a trained coronary heart disease recurrence risk scoring model to obtain a coronary heart disease recurrence risk score of the target patient.

[0005] In a possible implementation, the dynamic behavior data includes the actual number of times of taking medicine and the actual number of completed examination items; calculating the treatment compliance indicator of the target patient based on the dynamic behavior data and the physiological data of the target patient, comprising: calculating the ratio of the actual number of times of taking medicine of the target patient to the number of times of taking medicine, to obtain the medicine compliance indicator of the target patient; calculating the ratio of the actual number of completed examination items of the target patient to the number of completed examination items, to obtain the examination compliance indicator of the target patient; The medication compliance index and the examination compliance index are weighted and summed based on preset weights, to obtain the treatment compliance index of the target patient.

[0006] In a possible implementation, the dynamic behavior data includes exercise duration and exercise heart rate, and the physiological data includes ST segment deviation; and a calculation formula of the rehabilitation health index is:

[0007] wherein, is the rehabilitation health index, is the exercise duration, is the exercise heart rate, is a standardization factor, is the ST segment deviation.

[0008] In a possible implementation, the treatment compliance index, the rehabilitation health index and the static risk index are input into a trained coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient, including: The treatment compliance index, the rehabilitation health index and the static risk index are weighted and summed to obtain the coronary heart disease recurrence risk score of the target patient.

[0009] In a possible implementation, before the treatment compliance index, the rehabilitation health index and the static risk index are input into the coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient, further including: Obtaining individual characteristic data of the target patient; wherein the individual characteristic data includes age and myocardial infarction history; Adjusting weights of indexes in the coronary heart disease recurrence risk scoring model based on the age and the myocardial infarction history of the target patient to obtain an optimized coronary heart disease recurrence risk scoring model; Correspondingly, the treatment compliance index, the rehabilitation health index and the static risk index are input into the coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient, including: The treatment compliance index, the rehabilitation health index and the static risk index are input into the optimized coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient.

[0010] In a possible implementation, before the weights of indexes in the coronary heart disease recurrence risk scoring model are adjusted based on the age and the myocardial infarction history of the target patient to obtain the optimized coronary heart disease recurrence risk scoring model, further including: obtaining a historical patient data set, wherein the historical patient data set comprises a plurality of historical patient data in a plurality of annual interval data after the first onset; each annual interval data comprises: a start time and an end time of the interval, an identification of whether to relapse, individual characteristic data of the historical patient in the interval, dynamic behavior and physiological data; training a time-dependent Cox proportional hazards model based on the historical patient data set by using a survival analysis algorithm to obtain a trained Cox proportional hazards model; wherein the dynamic behavior and physiological data are used as time-dependent covariates, the individual characteristic data are used as fixed covariates, and at least one interaction term of the individual characteristic data and the dynamic behavior or physiological data is introduced into the Cox proportional hazards model; Correspondingly, based on the age and the history of myocardial infarction of the target patient, the weights of each index in the coronary heart disease recurrence risk score model are adjusted to obtain an optimized coronary heart disease recurrence risk score model, comprising: Based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model and the age and the history of myocardial infarction of the target patient, the weights of each index in the coronary heart disease recurrence risk score model are calculated.

[0011] In one possible implementation, based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model and the age and the history of myocardial infarction of the target patient, the formula for calculating the weights of each index in the coronary heart disease recurrence risk score model is:

[0012] wherein, is the weight coefficient of the index is the weight coefficient of the index is the treatment compliance index, the rehabilitation health index or the static risk index, is the main effect coefficient of the index in the trained Cox proportional hazards model, is the interaction term coefficient of the history of myocardial infarction and the index in the trained Cox proportional hazards model, is a binary variable, which is 1 when the target patient has a history of myocardial infarction, and 0 otherwise, is the interaction term coefficient of the age and the index in the trained Cox proportional hazards model, is the age of the target user, is the reference age.

[0013] In a second aspect, an embodiment of the present application provides a coronary heart disease recurrence risk assessment device, comprising: The acquisition module is configured to obtain dynamic behavior data, physiological data and static risk indexes of a target patient. a calculation module, configured to calculate a treatment adherence index and a rehabilitation health index of the target patient based on dynamic behavior data and physiological data of the target patient; a scoring module, configured to input the treatment adherence index, the rehabilitation health index and the static risk index into a trained coronary heart disease recurrence risk scoring model to obtain a coronary heart disease recurrence risk score of the target patient.

[0014] In a third aspect, an electronic device is provided, including a memory and a processor, the memory storing a computer program, and the processor implementing the method in the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0015] In a fourth aspect, a computer readable storage medium is provided, storing a computer program, and the computer program is executed by a processor to implement the method in the first aspect or any possible implementation manner of the first aspect.

[0016] The coronary heart disease recurrence risk assessment method, device, equipment and storage medium provided by the embodiments of the present application realize multi-dimensional and all-round assessment of coronary heart disease recurrence risk by obtaining dynamic behavior data, physiological data and static risk indicators of the target patient, effectively overcoming the limitations of relying only on static medical indicators in the prior art. By quantifying the dynamic behavior data and physiological data into the treatment adherence index and the rehabilitation health index, the behavior adherence and physiological state changes of the patient in daily medication, examination, exercise and the like are included in the assessment system, which significantly improves the comprehensiveness and real-time performance of the assessment. Finally, by inputting the above indicators into the trained coronary heart disease recurrence risk scoring model for comprehensive analysis, an objective and quantitative recurrence risk score can be output, providing more accurate decision support for clinicians, which helps to realize early warning and personalized intervention, thereby improving patient prognosis and reducing medical costs. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is an implementation flowchart of the coronary heart disease recurrence risk assessment method provided by the embodiments of the present application; Figure 2 is a structural schematic diagram of the coronary heart disease recurrence risk assessment device provided by the embodiments of the present application; Figure 3 is a schematic diagram of the electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0018] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0019] Reference is made to Figure 1 which shows an implementation flowchart of the coronary heart disease recurrence risk assessment method provided by the embodiments of the present application, and is described in detail as follows: Step 101, obtaining target patient dynamic behavior data, physiological data and static risk indicators.

[0020] In this embodiment, it needs to be pointed out that "recurrence" specifically refers to the occurrence of acute exacerbation or the appearance of new ischemic clinical events in the course of chronic coronary heart disease (a progressive and incurable atherosclerotic disease). This definition is based on the pathological nature of the disease - the possible continuous progression or sudden rupture of atherosclerotic plaques, leading to acute clinical symptoms.

[0021] Specifically, "recurrence events" can include the following situations: Acute coronary syndrome: including unstable angina, non-ST segment elevation myocardial infarction or ST segment elevation myocardial infarction; rehospitalization due to worsening of coronary heart disease symptoms: such as significant worsening of stable angina symptoms, or the appearance of serious arrhythmia, heart failure and other complications of coronary heart disease requiring hospitalization; Need for revascularization treatment: first or again receiving percutaneous coronary intervention or coronary artery bypass grafting after assessment during the follow-up period.

[0022] This program focuses on the state transition between the stable and acute exacerbation stages of the disease, which is the core goal of this risk assessment program.

[0023] This program is designed for patients in the long-term management stage of chronic coronary heart disease, and the target population includes patients who have been diagnosed as coronary heart disease (including stable angina, past myocardial infarction, or have received revascularization treatment); patients in the post-discharge rehabilitation period and regular outpatient follow-up stage; patients who need long-term secondary prevention and risk stratification management.

[0024] The core application scenarios include post-discharge risk reevaluation: after the patient is discharged after an acute event (such as 3 months, 6 months, 1 year), the long-term risk of the patient is dynamically and individually evaluated using this program to identify high-risk individuals; regular outpatient follow-up management: as part of routine outpatient review, integrate the patient's recent behavior and physiological data to quickly assess the current risk status and provide a basis for adjusting the treatment plan; rehabilitation effect monitoring: by continuously evaluating the changes in rehabilitation health index, the effectiveness and safety of rehabilitation training are quantified to guide the individual adjustment of rehabilitation plans.

[0025] This program is particularly suitable for identifying "hidden" high-risk patients who have acceptable static indicators but poor behavior compliance or potential physiological deterioration, thereby achieving early warning and intervention.

[0026] During the data acquisition phase, the system collects dynamic behavioral data, physiological data, and static risk indicators from target patients using various technologies. Dynamic behavioral data refers to the patient's behavior over time during recovery, primarily including medication timestamps recorded by smart pillboxes, examination completion status obtained through the hospital appointment system, and exercise duration and intensity data collected by wearable devices. Physiological data consists of patient physiological status-related monitoring data acquired through medical-grade equipment, including ST segment deviation from Holter monitoring and blood pressure variability recorded by a home blood pressure monitor. Static risk indicators are relatively stable baseline health indicators for the patient, which can be directly extracted from the hospital information system, including the most recently measured LDL cholesterol level, blood pressure, and coronary CT angiography stenosis score.

[0027] These data are transmitted and integrated through standardized application programming interfaces (APIs) to ensure data integrity and timeliness.

[0028] Step 102: Calculate the treatment compliance index and rehabilitation health index of the target patient based on the dynamic behavioral data and physiological data of the target patient.

[0029] In this embodiment, during the indicator calculation phase, the system preprocesses and extracts features from the collected raw data. Treatment adherence indicators reflect the degree to which patients follow medical advice, including medication adherence and examination adherence. The calculation of treatment adherence indicators integrates two dimensions: medication adherence and examination adherence. Medication adherence is determined by comparing the actual number of medication doses with the number required by the medical advice, while examination adherence is calculated by statistically analyzing the degree to which the actual number of completed examinations matches the required number of examinations.

[0030] The Rehabilitation Health Index is a quantitative indicator that integrates exercise behavior and myocardial function status to assess the rehabilitation effect of patients. The calculation of the Rehabilitation Health Index combines exercise data and physiological indicators, taking into account not only exercise duration and intensity, but also incorporating physiological parameters such as ST segment deviation as a moderating factor to ensure that the assessment results reflect both rehabilitation effectiveness and safety.

[0031] The risk of coronary heart disease recurrence is influenced by the patient's treatment compliance, rehabilitation status, and underlying health conditions. This embodiment integrates dynamic behavioral data (reflecting treatment adherence), physiological data (reflecting bodily functions), and static risk indicators (reflecting underlying risk), and through quantitative analysis and model calculation, achieves a comprehensive risk assessment.

[0032] Step 103: Input the treatment adherence index, rehabilitation health index and static risk index into the trained coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient.

[0033] In this embodiment, treatment adherence indicators, rehabilitation health index, and static risk indicators (such as total cholesterol 5.2 mmol / L and hypertension history assigned a value of 1) are input into the trained coronary heart disease recurrence risk scoring model, and the risk score is output (such as 65 points, out of 100 points, the higher the score, the higher the risk).

[0034] The coronary artery disease recurrence risk scoring model is a machine learning model trained on a large amount of clinical data. This model, trained using supervised learning algorithms, can effectively identify the complex relationship between multi-dimensional features and recurrence risk. The model input includes the aforementioned calculated treatment adherence indicators, rehabilitation health index, and static risk indicators. The output is a standardized risk score of 0-100, with higher scores indicating a greater risk of recurrence. The entire assessment process is conducted while ensuring data security; all sensitive information is anonymized, and the assessment results are presented through a visual interface, providing an intuitive reference for clinical decision-making.

[0035] This invention achieves a multi-dimensional and comprehensive assessment of the risk of coronary heart disease recurrence by acquiring dynamic behavioral data, physiological data, and static risk indicators of target patients, effectively overcoming the limitations of existing technologies that rely solely on static medical indicators. By quantifying dynamic behavioral and physiological data into treatment adherence indicators and rehabilitation health indices, and incorporating patients' daily medication adherence, examination, exercise, and other behavioral adherence and physiological state changes into the assessment system, the comprehensiveness and real-time nature of the assessment are significantly improved. Finally, by inputting the above indicators into a trained coronary heart disease recurrence risk scoring model for comprehensive analysis, an objective and quantitative recurrence risk score can be output, providing clinicians with more accurate decision support, facilitating early warning and personalized intervention, thereby improving patient prognosis and reducing medical costs.

[0036] In one possible implementation, dynamic behavioral data includes the actual number of medication doses and the actual number of examinations completed; treatment adherence indicators for the target patient are calculated based on the target patient's dynamic behavioral and physiological data, including: The ratio of the actual number of times a target patient takes medication to the number of times they should take medication is calculated to obtain the medication adherence index for the target patient. The ratio of the number of examinations actually completed to the number of examinations that should have been completed by the target patient is calculated to obtain the examination compliance index of the target patient; Based on preset weights, the medication adherence index and the examination adherence index are weighted and summed to obtain the treatment adherence index of the target patient.

[0037] In this embodiment, in actual implementation, medication adherence indicators are calculated using smart pillboxes or patient-reported medication records. Specifically, the system sets a statistical period (usually 30 days), records the actual number of times the patient takes medication within this period, and compares it with the number of times required by the doctor's prescription. The calculation of examination adherence indicators is more complex, requiring the integration of appointment records from the hospital information system and execution records from the laboratory. The system automatically compares the examination items required by the doctor's prescription with the actual completed examination items, taking into account differences in the importance and timeliness of the examination items.

[0038] In terms of weighting, this invention employs a weighting strategy based on clinical evidence. Medication adherence is given a high weight (typically 0.6-0.7) because regular medication has been proven to be the most critical factor in preventing relapse; check-up adherence has a relatively low weight (typically 0.3-0.4) because its direct impact on relapse risk is weaker but still has important reference value. This weighting scheme is derived through regression analysis of a large amount of clinical follow-up data, ensuring that the final calculated treatment adherence index has the best predictive power. During the calculation process, the system also considers the completeness and quality of the data, appropriately handling missing data to ensure the reliability of the evaluation results.

[0039] In one possible implementation, dynamic behavioral data includes exercise duration and exercise heart rate, and physiological data includes ST segment shift; the formula for calculating the rehabilitation health index is:

[0040] in, For the recovery health index, For exercise duration, For exercise heart rate, For standardization factors, This is the offset of the ST segment.

[0041] In this embodiment, exercise duration (ET) is determined by combining data from the wearable device's accelerometer and heart rate sensor, and only the effective exercise time reaching moderate intensity or above is counted. Exercise heart rate (HR) is taken from the average heart rate value during exercise, obtained through continuous monitoring by the sensors. ST segment offset (STO) is extracted from Holter monitoring data, using the most significant offset amplitude value.

[0042] The calculation formula is designed based on physiological principles and clinical evidence. The numerator (ET×HR) comprehensively reflects exercise volume and intensity, while the denominator... This is an empirically standardized factor used to adjust the values ​​to a reasonable range. It can be set and dynamically adjusted based on historical data. ST segment offset, as a negative adjustment factor, is designed so that the rehabilitation health index decreases accordingly when significant myocardial ischemia is present, consistent with clinical safety considerations. This formula has been clinically validated, effectively distinguishing patients with different rehabilitation conditions, and shows good consistency with traditional exercise stress test results.

[0043] In one possible implementation, treatment adherence indicators, rehabilitation health index, and static risk indicators are input into a trained coronary artery disease recurrence risk scoring model to obtain a coronary artery disease recurrence risk score for the target patient, including: The coronary artery disease recurrence risk score of the target patients was obtained by weighted summation of treatment adherence indicators, rehabilitation health index, and static risk indicators.

[0044] In this embodiment, the coronary artery disease recurrence risk scoring model can be a weighted summation risk scoring calculation model. The weight allocation of each indicator is obtained by training a historical dataset using a machine learning algorithm. The training data contains a large number of complete follow-up records of patients diagnosed with coronary artery disease, including their dynamic behavioral data, physiological indicators, static risk indicators, and endpoint data on whether a recurrence event occurred.

[0045] The model employs a linear weighted summation architecture, a design that ensures both computational efficiency and interpretability. Each indicator undergoes standardization before being input into the model to eliminate the influence of dimensional differences. The standardization method used is z-score standardization, which involves subtracting the mean of each indicator from its training set value and then dividing by the standard deviation. This process ensures comparability between different indicators and adapts to the characteristic distributions of different populations. The model's final output is transformed using a sigmoid function, mapping the linear weighted sum to a risk score range of 0-100 for easy clinical interpretation and application.

[0046] In one possible implementation, before inputting treatment adherence indicators, rehabilitation health index, and static risk indicators into the coronary artery disease recurrence risk scoring model to obtain the target patient's coronary artery disease recurrence risk score, the following steps are also included: Obtain individual characteristic data of the target patients; among which, individual characteristic data includes age and history of myocardial infarction; The weights of each indicator in the coronary heart disease recurrence risk scoring model were adjusted based on the age and myocardial infarction history of the target patients to obtain the optimized coronary heart disease recurrence risk scoring model. Accordingly, treatment adherence indicators, rehabilitation health index, and static risk indicators are input into the coronary artery disease recurrence risk scoring model to obtain the coronary artery disease recurrence risk score for the target patient, including: By inputting treatment adherence indicators, rehabilitation health index, and static risk indicators into the optimized coronary artery disease recurrence risk scoring model, the coronary artery disease recurrence risk score of the target patient is obtained.

[0047] In this embodiment, the patient's age and history of myocardial infarction are first obtained through an electronic medical record system. Age information is required to be accurate to a specific numerical value, while the history of myocardial infarction is recorded as a binary variable (yes / no). These individual characteristic data are cross-validated with diagnostic records in the clinical database to ensure accuracy.

[0048] Weight adjustment employs a combination of rule-based engine and machine learning. For age factors, the system sets 65 years as a threshold; when a patient's age exceeds this threshold, the weight of static risk indicators automatically increases by 5-10 percentage points, as older patients are more sensitive to traditional risk factors. For patients with a history of myocardial infarction, the weight of the rehabilitation health index increases by 8-12 percentage points, reflecting the specific requirements of this population for rehabilitation quality. After weight adjustment, the system re-normalizes to ensure the sum of the weights of all indicators remains 1. This dynamic adjustment mechanism significantly improves the model's predictive accuracy across different patient subgroups.

[0049] In one possible implementation, before adjusting the weights of various indicators in the coronary artery disease recurrence risk scoring model based on the target patient's age and history of myocardial infarction to obtain the optimized coronary artery disease recurrence risk scoring model, the following steps are also included: Obtain a historical patient dataset; the historical patient dataset includes data from multiple historical patients over multiple year intervals after the first onset of illness; each year interval includes: the start and end time of the interval, a relapse indicator, the individual characteristics of the historical patient in that interval, and dynamic behavioral and physiological data; Based on historical patient datasets, a time-dependent Cox proportional hazards model was trained using a survival analysis algorithm to obtain a trained Cox proportional hazards model. Dynamic behavior and physiological data were used as time-dependent covariates, individual characteristic data were used as fixed covariates, and at least one interaction term between individual characteristic data and dynamic behavior or physiological data was introduced into the Cox proportional hazards model. Accordingly, the weights of each indicator in the coronary artery disease recurrence risk scoring model are adjusted based on the target patient's age and history of myocardial infarction, resulting in an optimized coronary artery disease recurrence risk scoring model, including: Based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model, as well as the age and myocardial infarction history of the target patient, the weights of each indicator in the coronary heart disease recurrence risk scoring model are calculated.

[0050] In this embodiment, constructing the historical dataset is crucial, requiring the collection of at least three years of patient follow-up data. Each patient's data is divided into multiple observation intervals by year, with each interval containing behavioral data, physiological indicators, static risk indicators, and endpoint information regarding whether a relapse event occurred for that year. Data quality control includes steps such as outlier handling, missing value imputation, and consistency verification.

[0051] The model training employed a time-dependent Cox proportional hazards model, which effectively handles right-censored data and is suitable for medical follow-up studies. Dynamic behavioral and physiological data were used as time-dependent covariate inputs, the values ​​of which vary across observation intervals. Individual characteristic data were used as fixed covariate inputs. The model intentionally introduced interaction terms, including interactions between age and static risk indicators, and between myocardial infarction history and rehabilitation health index. These interactions capture the differential effects of risk factors among different patient subgroups. The model training used the maximum likelihood estimation method, employing the Breslow method to process the final data, ensuring the accuracy of coefficient estimation.

[0052] In one possible implementation, based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model, as well as the target patient's age and history of myocardial infarction, the formula for calculating the weights of each indicator in the coronary artery disease recurrence risk scoring model is as follows:

[0053] in, As an indicator The weighting coefficients, For treatment adherence indicators, rehabilitation health index, or static risk indicators, Indicators in the trained Cox proportional hazards model The main effect coefficient, For the completed Cox proportional hazards model, the history and indicators of central infarction were analyzed. The interaction term coefficients, This is a binary variable; it takes a value of 1 if the target patient has a history of myocardial infarction, and 0 otherwise. For the completed Cox proportional hazards model, age and indicators The interaction term coefficients, For the target user's age, The baseline age is used.

[0054] In this embodiment, the main effect coefficient in the formula The interaction term coefficient reflects the independent impact of each indicator on the risk of relapse. and This quantifies the moderating effect of individual characteristics on indicator weights. Baseline age The average age of the training population (e.g., 65 years old) is usually used to standardize the age effect.

[0055] In the specific calculations, the system first extracts the required coefficient matrix from the trained Cox model. For each patient to be assessed, the personalized weights of each indicator are dynamically calculated based on their specific age and history of myocardial infarction. For example, the weight calculation of static risk indicators considers their main effect coefficient, the interaction term coefficient with myocardial infarction history, and the interaction term coefficient with age. This calculation method ensures the scientific rigor of individualized risk assessment, respecting statistical laws while taking into account clinical practice. Finally, the calculated weights are processed by a softmax function to ensure that all weights are positive and sum to 1, meeting the requirements of the weighted summation model.

[0056] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0057] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0058] Figure 2 A schematic diagram of the coronary heart disease recurrence risk assessment device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the coronary heart disease recurrence risk assessment device 2 includes: Module 21 is used to acquire dynamic behavioral data, physiological data and static risk indicators of the target patient; Calculation module 22 is used to calculate the treatment compliance index and rehabilitation health index of the target patient based on the target patient's dynamic behavioral data and physiological data; The scoring module 23 is used to input treatment compliance indicators, rehabilitation health index and static risk indicators into the trained coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient.

[0059] In one possible implementation, the dynamic behavioral data includes the actual number of medication doses and the actual number of completed examinations; the calculation module 22 is specifically used for: The ratio of the actual number of times a target patient takes medication to the number of times they should take medication is calculated to obtain the medication adherence index for the target patient. The ratio of the number of examinations actually completed to the number of examinations that should have been completed by the target patient is calculated to obtain the examination compliance index of the target patient; Based on preset weights, the medication adherence index and the examination adherence index are weighted and summed to obtain the treatment adherence index of the target patient.

[0060] In one possible implementation, dynamic behavioral data includes exercise duration and exercise heart rate, and physiological data includes ST segment shift; the formula for calculating the rehabilitation health index is:

[0061] in, For the recovery health index, For exercise duration, For exercise heart rate, For standardization factors, This is the offset of the ST segment.

[0062] In one possible implementation, the scoring module 23 is specifically used for: The coronary artery disease recurrence risk score of the target patients was obtained by weighted summation of treatment adherence indicators, rehabilitation health index, and static risk indicators.

[0063] In one possible implementation, the scoring module 23 is also used for: Before inputting treatment adherence indicators, rehabilitation health index, and static risk indicators into the coronary artery disease recurrence risk scoring model to obtain the coronary artery disease recurrence risk score of the target patient, individual characteristic data of the target patient is obtained; among which, individual characteristic data includes age and history of myocardial infarction; The weights of each indicator in the coronary heart disease recurrence risk scoring model were adjusted based on the age and myocardial infarction history of the target patients to obtain the optimized coronary heart disease recurrence risk scoring model. Accordingly, treatment adherence indicators, rehabilitation health index, and static risk indicators are input into the coronary artery disease recurrence risk scoring model to obtain the coronary artery disease recurrence risk score for the target patient, including: By inputting treatment adherence indicators, rehabilitation health index, and static risk indicators into the optimized coronary artery disease recurrence risk scoring model, the coronary artery disease recurrence risk score of the target patient is obtained.

[0064] In one possible implementation, the scoring module 23 is also used for: Before adjusting the weights of various indicators in the coronary artery disease recurrence risk scoring model based on the target patient's age and history of myocardial infarction to obtain the optimized coronary artery disease recurrence risk scoring model, a historical patient dataset is obtained. The historical patient dataset includes data from multiple historical patients over multiple year intervals after their first onset. Each year interval includes: the start and end time of the interval, the indicator of whether a recurrence has occurred, the individual characteristics data of the historical patient in that interval, and dynamic behavioral and physiological data. Based on historical patient datasets, a time-dependent Cox proportional hazards model was trained using a survival analysis algorithm to obtain a trained Cox proportional hazards model. Dynamic behavior and physiological data were used as time-dependent covariates, individual characteristic data were used as fixed covariates, and at least one interaction term between individual characteristic data and dynamic behavior or physiological data was introduced into the Cox proportional hazards model. Accordingly, the weights of each indicator in the coronary artery disease recurrence risk scoring model are adjusted based on the target patient's age and history of myocardial infarction, resulting in an optimized coronary artery disease recurrence risk scoring model, including: Based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model, as well as the age and myocardial infarction history of the target patient, the weights of each indicator in the coronary heart disease recurrence risk scoring model are calculated.

[0065] In one possible implementation, based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model, as well as the target patient's age and history of myocardial infarction, the formula for calculating the weights of each indicator in the coronary artery disease recurrence risk scoring model is as follows:

[0066] in, As an indicator The weighting coefficients, For treatment adherence indicators, rehabilitation health index, or static risk indicators, Indicators in the trained Cox proportional hazards model The main effect coefficient, For the completed Cox proportional hazards model, the history and indicators of central infarction were analyzed. The interaction term coefficients, This is a binary variable; it takes a value of 1 if the target patient has a history of myocardial infarction, and 0 otherwise. For the completed Cox proportional hazards model, age and indicators The interaction term coefficients, For the target user's age, The baseline age is used.

[0067] This invention achieves a multi-dimensional and comprehensive assessment of the risk of coronary heart disease recurrence by acquiring dynamic behavioral data, physiological data, and static risk indicators of target patients, effectively overcoming the limitations of existing technologies that rely solely on static medical indicators. By quantifying dynamic behavioral and physiological data into treatment adherence indicators and rehabilitation health indices, and incorporating patients' daily medication adherence, examination, exercise, and other behavioral adherence and physiological state changes into the assessment system, the comprehensiveness and real-time nature of the assessment are significantly improved. Finally, by inputting the above indicators into a trained coronary heart disease recurrence risk scoring model for comprehensive analysis, an objective and quantitative recurrence risk score can be output, providing clinicians with more accurate decision support, facilitating early warning and personalized intervention, thereby improving patient prognosis and reducing medical costs.

[0068] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Figure 3 As shown, the electronic device 3 of this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the various method embodiments described above. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module / unit in the various device embodiments described above.

[0069] For example, computer program 32 may be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 32 in electronic device 3.

[0070] Electronic device 3 may include, but is not limited to, processor 30 and memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 3 may also include input / output devices, network access devices, buses, etc.

[0071] The processor 30 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0072] The memory 31 can be an internal storage unit of the electronic device 3, such as a hard disk or memory of the electronic device 3. The memory 31 can also be an external storage device of the electronic device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 3. Furthermore, the memory 31 can include both internal and external storage units of the electronic device 3. The memory 31 is used to store the computer program 32 and other programs and data required by the electronic device 3. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0073] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0074] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0075] This invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0076] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0077] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0078] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for assessing the risk of recurrence of coronary heart disease, characterized in that, include: Acquire dynamic behavioral data, physiological data, and static risk indicators of the target patients; The treatment compliance index and the rehabilitation health index of the target patient are calculated based on the dynamic behavioral data and physiological data of the target patient. The treatment adherence index, the rehabilitation health index, and the static risk index are input into a trained coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score for the target patient.

2. The method for assessing the risk of coronary heart disease recurrence according to claim 1, characterized in that, The dynamic behavioral data includes the actual number of times medication was taken and the actual number of examinations completed. The treatment adherence indicators of the target patients are calculated based on their dynamic behavioral and physiological data, including: The ratio of the actual number of times the target patient takes medication to the number of times the medication should be taken is calculated to obtain the medication adherence index of the target patient; The ratio of the number of examinations actually completed to the number of examinations that should have been completed by the target patient is calculated to obtain the examination compliance index of the target patient; Based on preset weights, the medication adherence index and the examination adherence index are weighted and summed to obtain the treatment adherence index of the target patient.

3. The method for assessing the risk of coronary heart disease recurrence according to claim 1, characterized in that, The dynamic behavioral data includes exercise duration and exercise heart rate, and the physiological data includes ST segment deviation; the formula for calculating the rehabilitation health index is: in, For the recovery health index, For exercise duration, For exercise heart rate, For standardization factors, This is the offset of the ST segment.

4. The method for assessing the risk of coronary heart disease recurrence according to claim 1, characterized in that, The step involves inputting the treatment adherence index, the rehabilitation health index, and the static risk index into a trained coronary artery disease recurrence risk scoring model to obtain the coronary artery disease recurrence risk score for the target patient, including: The treatment adherence index, the rehabilitation health index, and the static risk index are weighted and summed to obtain the coronary heart disease recurrence risk score of the target patient.

5. The method for assessing the risk of coronary heart disease recurrence according to claim 4, characterized in that, Before inputting the treatment adherence index, the rehabilitation health index, and the static risk index into the coronary artery disease recurrence risk scoring model to obtain the coronary artery disease recurrence risk score of the target patient, the method further includes: Obtain individual characteristic data of the target patient; wherein, the individual characteristic data includes age and history of myocardial infarction; The weights of each indicator in the coronary heart disease recurrence risk scoring model were adjusted based on the age and myocardial infarction history of the target patients to obtain the optimized coronary heart disease recurrence risk scoring model. Accordingly, the step of inputting the treatment adherence index, the rehabilitation health index, and the static risk index into the coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient includes: The treatment adherence index, the rehabilitation health index, and the static risk index are input into the optimized coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient.

6. The method for assessing the risk of coronary heart disease recurrence according to claim 5, characterized in that, Before adjusting the weights of each indicator in the coronary artery disease recurrence risk scoring model based on the target patient's age and history of myocardial infarction to obtain the optimized coronary artery disease recurrence risk scoring model, the following steps are also included: Obtain a historical patient dataset; wherein the historical patient dataset includes data from multiple historical patients over multiple year intervals after the first onset of illness; each year interval includes: the start and end time of the interval, a relapse indicator, individual characteristic data of the historical patient in the interval, and dynamic behavioral and physiological data; Based on the historical patient dataset, a time-dependent Cox proportional hazards model is trained using a survival analysis algorithm to obtain a trained Cox proportional hazards model. The dynamic behavior and physiological data are used as time-dependent covariates, the individual characteristic data are used as fixed covariates, and at least one interaction term between the individual characteristic data and the dynamic behavior or physiological data is introduced into the Cox proportional hazards model. Accordingly, the optimized coronary heart disease recurrence risk scoring model is obtained by adjusting the weights of each indicator in the model based on the target patient's age and history of myocardial infarction, including: Based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model, as well as the age and myocardial infarction history of the target patient, the weights of each indicator in the coronary heart disease recurrence risk scoring model are calculated.

7. The method for assessing the risk of coronary heart disease recurrence according to claim 6, characterized in that, The formula for calculating the weights of each indicator in the coronary artery disease recurrence risk scoring model, based on the coefficients of each covariate and interaction term in the trained Cox proportional hazards model, and the age and myocardial infarction history of the target patient, is as follows: in, As an indicator The weighting coefficients, These are the treatment adherence indicators, the rehabilitation health index, or the static risk indicators. Indicators in the completed Cox proportional hazards model The main effect coefficient, For the completed Cox proportional hazards model, the history and indicators of central infarction were analyzed. The interaction term coefficients, This is a binary variable; it takes a value of 1 if the target patient has a history of myocardial infarction, and 0 otherwise. For the completed Cox proportional hazards model, age and indicators The interaction term coefficients, The age of the target user, The baseline age is used.

8. A device for assessing the risk of coronary heart disease recurrence, characterized in that, include: The acquisition module is used to acquire dynamic behavioral data, physiological data, and static risk indicators of the target patient. The calculation module is used to calculate the treatment compliance index and the rehabilitation health index of the target patient based on the target patient's dynamic behavioral data and physiological data; The scoring module is used to input the treatment compliance index, the rehabilitation health index and the static risk index into the trained coronary heart disease recurrence risk scoring model to obtain the coronary heart disease recurrence risk score of the target patient.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images