Atherosclerotic event risk prediction method, apparatus, device, and medium
By constructing a multivariate Cox proportional hazards model for metabolic inflammatory liver indices and hepatic vascular protective biomarkers, the problem of insufficient identification of high-risk individuals in the intermediate-risk group by existing models is solved, and efficient risk assessment and prediction of atherosclerotic events is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FUWAI HOSPITAL CHINESE ACAD OF MEDICAL SCI & PEKING UNION MEDICAL COLLEGE
- Filing Date
- 2026-04-10
- Publication Date
- 2026-07-10
AI Technical Summary
Existing atherosclerosis risk prediction models are not sensitive enough in identifying high-risk individuals in the intermediate-risk group, resulting in missed prevention opportunities. Traditional risk factors have limited predictive capabilities, and there is an urgent need to integrate novel biomarkers to improve the accuracy of risk stratification.
We constructed a metabolic inflammatory liver index and a hepatic vascular protective biomarker index, combined with traditional atherosclerosis risk variables, and adopted a multivariate Cox proportional hazards model. Variables were screened through univariate Cox regression and LASSO-Cox regression strategies to establish a risk prediction model for atherosclerosis events.
It significantly improves the ability to identify medium- and long-term atherosclerotic events, provides a new risk assessment tool, can more accurately identify high-risk groups, and supports effective primary prevention.
Smart Images
Figure CN122369923A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, device, and medium for predicting the risk of atherosclerotic events. Background Technology
[0002] Atherosclerosis is the main pathological basis of cardiovascular and cerebrovascular events, and early identification of high-risk groups is crucial for primary prevention. Traditional risk factors have limited predictive capabilities, and there is an urgent need to integrate novel biomarkers to improve the accuracy of risk stratification.
[0003] Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death and disability worldwide. Its pathological process is lengthy and insidious, making early identification of high-risk groups crucial for effective primary prevention. [1] Currently, clinical practice widely relies on predictive models (such as Pooled Cohort Equations, QRISK3) based on traditional risk factors (such as age, sex, blood pressure, blood lipids, smoking, and diabetes) for risk assessment. ] However, these models have significant limitations in identifying actual high-risk individuals within the "medium-risk" group, and their discrimination is often insufficient, resulting in a large number of missed prevention opportunities. Summary of the Invention
[0004] To address the problems existing in the prior art, the present invention provides a method, device, equipment, and medium for predicting the risk of atherosclerotic events.
[0005] This invention provides a method for predicting the risk of atherosclerotic events, comprising: Baseline clinical data was obtained, and metabolic inflammatory liver indices and hepatic vascular protective biomarker indices were determined based on the baseline clinical data; the baseline clinical data included traditional atherosclerosis risk variables; Based on the aforementioned metabolic inflammatory liver index, hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, a multivariate Cox proportional hazards model was established. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the corresponding risk of atherosclerotic events is output.
[0006] According to the present invention, a method for predicting the risk of atherosclerotic events is provided, wherein a multivariate Cox proportional hazards prediction model is established based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index, and traditional atherosclerotic risk variables, including: A univariate Cox regression strategy was used to select candidate variables related to atherosclerotic events from traditional atherosclerosis risk variables; The candidate variables were screened using the LASSO-Cox regression strategy, and redundant variables were removed to obtain the variables used in the model. The model was trained based on the selected input variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index. The sensitivity and goodness of fit of the model were evaluated using time-dependent ROC curves, C-index, and AIC. When the sensitivity and evaluation values met the conditions, a multivariate Cox proportional hazards prediction model was obtained.
[0007] According to the present invention, a method for predicting the risk of atherosclerotic events is provided, wherein the model is trained based on the selected input variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index; the sensitivity and goodness of fit of the model are evaluated using time-dependent ROC curves, C-index, and AIC; and a multivariate Cox proportional hazards prediction model is obtained when the sensitivity and evaluation values meet the conditions, including: The input variables, including the selected variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index, were used as input variables. Atherosclerosis-related events were used as endpoint events, and follow-up time was used as the time variable. A multivariate Cox proportional hazards regression method was used to fit the model, and the regression coefficients of each variable were obtained. An atherosclerosis risk prediction model was constructed, and the sensitivity and goodness of fit of the model were evaluated to obtain a multivariate Cox proportional hazards prediction model that can quantify the individual's risk of developing the disease.
[0008] According to the method for predicting the risk of atherosclerotic events provided by the present invention, the metabolic inflammatory liver index is determined by body mass index (BMI), the ratio of lymphocytes to neutrophils, and C-reactive protein; the hepatic vascular protective biomarker index is determined by total bilirubin, albumin, alkaline phosphatase, and aspartate aminotransferase.
[0009] According to the present invention, a method for predicting the risk of atherosclerotic events is provided, wherein a univariate Cox regression strategy is used to select candidate variables related to atherosclerotic events from traditional atherosclerotic risk variables, including: Univariate Cox proportional hazards regression models were constructed for each traditional atherosclerosis risk variable; Using atherosclerotic events as the endpoint and follow-up time as the time scale, we fitted and calculated the significance values of each variable one by one, and selected the variables with significance values less than the preset values as candidate variables.
[0010] According to the present invention, a method for predicting the risk of atherosclerotic events is provided, wherein the LASSO-Cox regression strategy is used to screen the candidate variables and remove redundant variables to obtain the input variables, including: The LASSO-Cox regression model was used to screen candidate variables by penalized regression, and the optimal penalty coefficient λ was determined by ten-fold cross-validation. The variable coefficients are compressed based on the optimal penalty coefficient λ. Redundant variables whose coefficients are compressed to zero are removed, and variables with non-zero coefficients are retained as input variables.
[0011] According to the present invention, a method for predicting the risk of atherosclerotic events is provided, wherein the atherosclerotic events include one or more of myocardial infarction, ischemic stroke, and revascularization surgery.
[0012] The present invention also provides a device for predicting the risk of atherosclerotic events, comprising: The acquisition module is used to acquire baseline clinical data and determine metabolic inflammatory liver index and hepatic vascular protective biomarker index based on the baseline clinical data; the baseline clinical data includes traditional atherosclerosis risk variables; A module is established to build a multivariate Cox proportional hazards model based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index and traditional atherosclerosis risk variables. The prediction module is used to output the risk of occurrence of corresponding atherosclerotic events based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model.
[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement any of the above-described methods for predicting the risk of atherosclerotic events.
[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for predicting the risk of atherosclerotic events.
[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described methods for predicting the risk of atherosclerotic events.
[0016] This invention provides a method, device, equipment, and medium for predicting the risk of atherosclerotic events. It determines metabolic inflammatory liver indices and hepatic vascular protective biomarkers based on baseline clinical data. A multivariate Cox proportional hazards prediction model is established based on these indices and traditional atherosclerotic risk variables. The model outputs the corresponding risk of atherosclerotic events based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model. This significantly improves the ability to identify medium- and long-term risks, providing a new and potentially clinically valuable reference for risk assessment of atherosclerotic cardiovascular diseases. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in this 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0018] Figure 1 This is a flowchart illustrating the method for predicting the risk of atherosclerotic events provided by the present invention.
[0019] Figure 2 This is a schematic diagram showing the comparison between the training set and the external validation set in the ROC curve comparison provided by this invention.
[0020] Figure 3 This is a schematic diagram of finite cubic spline analysis of the risk score for atherosclerotic events provided by the present invention.
[0021] Figure 4 This is a schematic diagram of the joint distribution and risk contribution of MILI and HVPBI provided by the present invention.
[0022] Figure 5 This is a schematic diagram of the atherosclerotic event risk prediction device provided by the present invention.
[0023] Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation
[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0025] Figure 1 This invention provides a flowchart illustrating a method for predicting the risk of atherosclerotic events. (See attached diagram.) Figure 1 The method includes the following steps: Step 11: Obtain baseline clinical data and determine metabolic inflammatory liver index and hepatic vascular protective biomarker index based on the baseline clinical data; baseline clinical data includes traditional atherosclerosis risk variables.
[0026] Step 12: Based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, establish a multivariate Cox proportional hazards prediction model.
[0027] Step 13: Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, output the corresponding risk of atherosclerotic events.
[0028] Regarding steps 11-13, it should be noted that atherosclerosis is the main pathological basis of cardiovascular and cerebrovascular events, and early identification of high-risk groups is crucial for primary prevention. Traditional risk factors have limited predictive capabilities, necessitating the integration of novel biomarkers to improve the accuracy of risk stratification.
[0029] Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death and disability worldwide. Its pathological process is lengthy and insidious, making early identification of high-risk groups crucial for effective primary prevention. [1] Currently, clinical practice widely relies on predictive models (such as Pooled Cohort Equations, QRISK3) based on traditional risk factors (such as age, sex, blood pressure, blood lipids, smoking, and diabetes) for risk assessment. ] However, these models have significant limitations in identifying actual high-risk individuals within the "intermediate risk" group, often exhibiting insufficient sensitivity (discrimination) and resulting in missed preventative opportunities. Therefore, integrating novel biomarkers that more directly reflect the core pathophysiological processes of atherosclerosis to improve the accuracy of risk stratification has become a key direction in current cardiovascular prevention research.
[0030] The occurrence and development of atherosclerosis is far more complex than simple cholesterol deposition; it involves chronic low-grade inflammation, metabolic disorders, endothelial dysfunction, and an imbalance in tissue repair. In recent years, numerous studies have explored the predictive value of single novel biomarkers, such as C-reactive protein (CRP) and interleukin-6 (IL-6) reflecting inflammation, lipoprotein(a) reflecting abnormal lipid metabolism, and bilirubin and cystatin C reflecting liver and kidney function and oxidative stress. However, single biomarkers are easily affected by various confounding factors and can only reflect one aspect of the pathological process, resulting in limited improvement in predictive efficacy. Therefore, integrating multiple related biological indicators into a comprehensive index to more comprehensively capture an individual's overall pathophysiological state is becoming a promising strategy. For example, the liver, as a core organ for metabolic and inflammatory regulation, has its functional state closely related to vascular health through multiple mechanisms. Constructing a comprehensive index reflecting the liver's metabolic inflammatory burden and protective function may provide a new perspective for cardiovascular risk prediction.
[0031] To this end, this invention constructs two novel comprehensive biomarker indices—the metabolic inflammatory liver index and the hepatic vascular protective biomarker index. The metabolic inflammatory liver index is determined using body mass index (BMI), the lymphocyte-to-neutrophil ratio, and C-reactive protein; the hepatic vascular protective biomarker index is determined using total bilirubin, albumin, alkaline phosphatase, and aspartate aminotransferase.
[0032] In this invention, numerous data points can be collected about the human body, such as age, gender, race, education level, and family income. Lifestyle and behavioral factors include: smoking status (never, past, current), alcohol consumption frequency (never, occasionally, frequently), and physical activity level. Physical and physiological indicators include: body mass index (BMI), systolic blood pressure, diastolic blood pressure, and total body fat percentage.
[0033] Complete blood count: white blood cells, neutrophils, lymphocytes, monocytes, eosinophils, platelet count, and mean platelet volume, etc. Biochemistry and metabolism: cholesterol, glycated hemoglobin, cystatin C; Inflammatory markers: C-reactive protein, interleukin-6 (detected using the Olink platform, NPX value); Liver and kidney function and others: total bilirubin, albumin, alkaline phosphatase, aspartate aminotransferase, vitamin D, lipoprotein (a).
[0034] Medication usage: Use of lipid-lowering drugs and antihypertensive drugs.
[0035] Psychological factors: neuroticism score.
[0036] The above data serve as traditional risk variables for atherosclerosis, and these variables were collected using baseline clinical data.
[0037] To comprehensively assess metabolic inflammation and hepatic vascular protective function, this study constructed two standardized biomarker indices: (1) Metabolic inflammatory liver index (MILIraw) This index integrates obesity, inflammatory cell homeostasis, and systemic inflammatory load: MILIraw = Body Mass Index (BMI) × (Lymphocytes / Neutralocytes) × ln(C-reactive protein + 1) (2) Hepatic vascular protective biomarker index (HVPBIraw) This index integrates markers of liver protection (bilirubin, albumin) and damage (alkaline phosphatase, aspartate aminotransferase): HVPBIraw = [(Total bilirubin) × (Albumin)] / [(Alkaline phosphatase) × Aspartate aminotransferase)].
[0038] In this invention, the baseline clinical data is obtained by collecting detailed baseline information, biological samples, and long-term follow-up data from a large number of participants aged 40–69 years.
[0039] These data can be used as datasets to build and train models, ultimately resulting in a multivariate Cox proportional hazards model. This invention uses metabolic inflammatory liver indices, hepatic vascular protective biomarkers, and traditional atherosclerosis risk variables as data features to establish a multivariate Cox proportional hazards prediction model.
[0040] The multivariate Cox proportional hazards prediction model is a multifactor regression model used for survival analysis. It uses the event occurrence time and survival status as dependent variables and multiple influencing factors as independent variables. By constructing regression equations, it estimates the independent effects of each factor on the risk of event occurrence, eliminates confounding factors, and finally obtains the hazard ratio (HR), 95% confidence interval, and statistical significance of each variable.
[0041] The further method described above mainly explains the process of establishing a multivariate Cox proportional hazards prediction model based on metabolic inflammatory liver indices, hepatic vascular protective biomarkers, and traditional atherosclerosis risk variables, as detailed below: A univariate Cox regression strategy was used to select candidate variables related to atherosclerotic events from traditional atherosclerosis risk variables; The candidate variables were screened using the LASSO-Cox regression strategy, and redundant variables were removed to obtain the variables used in the model. The model was trained based on the selected input variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index. The sensitivity and goodness of fit of the model were evaluated using time-dependent ROC curves, C-index, and AIC. When the sensitivity and evaluation values met the conditions, a multivariate Cox proportional hazards prediction model was obtained.
[0042] It's important to note that while there are many traditional atherosclerosis risk variables, only a limited number are closely associated with atherosclerotic events; in other words, not all variables are included. Using weakly associated variables in model building and training could increase training difficulty and reduce predictive accuracy. Therefore, it's necessary to select variables closely related to the event from the large pool of variables and use them as candidate variables.
[0043] In this invention, a univariate Cox regression strategy is used to screen candidate variables. This mainly involves constructing univariate Cox proportional hazards regression models for each traditional atherosclerosis risk variable, using atherosclerotic events as the endpoint and follow-up time as the time scale, fitting the model one by one, calculating the significance value of each variable, and selecting variables with significance values less than a preset value as candidate variables.
[0044] Specifically: The outcome variable was atherosclerotic events, and the follow-up time was the time variable. Each candidate variable was individually substituted into the Cox proportional hazards regression model for univariate fitting. Calculate the hazard ratio (HR), 95% confidence interval, and significance level (P-value) for each variable; Variables with a significance value of P < 0.10 were identified as being related to the outcome and included in the candidate variable set for subsequent multifactor modeling and LASSO screening.
[0045] It should be noted that univariate Cox regression is only used to "screen" which variables are worth proceeding to the next step of modeling, and does not draw a final conclusion. Therefore, it is only necessary to use the p-value to determine whether the variable "has a statistical association". HR and 95% CI are not meaningful for screening at this time.
[0046] In this invention, redundant variables may exist among those closely related to the event, which is detrimental to model training and prediction accuracy. Therefore, it is necessary to further filter out some variables from the candidate variables, and the remaining variables are used as the input variables. In this invention, a LASSO-Cox regression strategy is used to screen the candidate variables and remove redundant variables to obtain the input variables. Specifically, the LASSO-Cox regression strategy involves using a LASSO-Cox regression model to perform penalized regression screening on the candidate variables, determining the optimal penalty coefficient λ through 10-fold cross-validation; based on the optimal penalty coefficient λ, the variable coefficients are compressed, eliminating redundant variables whose coefficients are compressed to zero, and retaining variables with non-zero coefficients as input variables.
[0047] Specifically: Input all candidate variables into the LASSO-Cox model; The model selects the optimal penalty parameter λ through cross-validation (CV); Automatically compress variable coefficients; Variables whose coefficients are compressed to 0 are used to eliminate redundant variables; Variables with non-zero coefficients are reserved as input variables; A further step in the above method mainly involves training the model based on the selected input variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index. The model's sensitivity and goodness of fit are evaluated using time-dependent ROC curves, C-index, and AIC. The process of obtaining the multivariate Cox proportional hazards prediction model when the sensitivity and evaluation values meet the conditions is explained below: The input variables, including the selected variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index, were used as input variables. Atherosclerosis-related events were used as endpoint events, and follow-up time was used as the time variable. A multivariate Cox proportional hazards regression method was used to fit the model, and the regression coefficients of each variable were obtained. An atherosclerosis risk prediction model was constructed, and the sensitivity and goodness of fit of the model were evaluated to obtain a multivariate Cox proportional hazards prediction model that can quantify the individual's risk of developing the disease.
[0048] To comprehensively evaluate the discriminative ability of the prediction model at different time scales, time-dependent ROC curves were plotted and the area under the curve (AUC) at the corresponding time points was calculated. See [link to relevant documentation]. Figure 2ROC curve comparison: A schematic diagram comparing the training set and the external validation set. The model demonstrated good predictive performance in the training cohort: its AUC values for predicting atherosclerotic events at 3 years, 5 years, and 10 years reached 0.776, 0.771, and 0.756, respectively. In the independent external validation cohort, the model also demonstrated robust and strong discriminative ability, with its AUC values for predicting events at 3 years, 5 years, and 10 years remaining robust at 0.787, 0.801, and 0.774, respectively. This result indicates that the model not only has accurate risk discrimination in the short term, but its predictive power remains stable over a follow-up period of up to ten years, and it has been effectively validated in an independent population, meeting the performance requirements of clinical long-term risk prediction tools.
[0049] In this invention, after obtaining the model, the risk of corresponding atherosclerotic events is output based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model. Atherosclerotic events include one or more of myocardial infarction, ischemic stroke, and revascularization surgery. The risk prediction is a medium- to long-term risk, including 3-year, 5-year, and 10-year incidence risks.
[0050] To explore the specific association between the two standardized scores, MILI and HVPBI, and the risk of atherosclerotic events, we conducted a dose-response analysis using a restricted cubic spline model. After adjusting for multiple confounding factors such as age, sex, blood pressure, smoking, diabetes, blood lipids, and inflammatory markers, the analysis results showed ( Figure 3 (This is a schematic diagram of a finite cubic spline analysis of the atherosclerotic event risk scores): The nonlinear p-value for the association between the MILI score and risk is 0.521, and the nonlinear p-value for the HVPBI score is 0.690, both significantly greater than the 0.05 significance level. This indicates that throughout the entire range of values observed in the study, both the MILI and HVPBI scores exhibit a statistically significant linear association with event risk, without revealing obvious "J-shaped" or "U-shaped" nonlinear trends. This linear relationship supports the rationale for treating both as continuous variables in the multivariate Cox regression model and suggests that their risk effect monotonically increases (MILI) or decreases (HVPBI) with the score value, facilitating intuitive risk assessment in clinical practice.
[0051] To further reveal the distribution patterns of the two risk biomarkers, MILI and HVPBI, in the population and their joint contribution to the final prediction model, a joint distribution map of the two was plotted, and the SHAP value was calculated based on Model 4 to quantify their risk contribution. Figure 4(A schematic diagram illustrating the joint distribution and risk contribution of MILI and HVPBI). This visualization clearly shows the risk stratification of the population: individuals are distributed in four characteristic quadrants based on their MILI and HVPBI scores. The clustering area of individuals with the "high MILI-low HVPBI" combination (represented by warm colors in the figure) corresponds to the highest SHAP contribution value, representing the highest disease risk in this subgroup; conversely, the "low MILI-high HVPBI" combination (represented by cool colors in the figure) is associated with the lowest risk. Synergistic and antagonistic effects of the indicators: The chart visually reveals the interaction effect between the two biomarkers in risk prediction. For example, a high level of HVPBI appears to partially offset the risk increment brought about by high MILI. This pattern suggests that simultaneously assessing MILI and HVPBI can achieve more accurate risk stratification than using either indicator alone. This also provides important clues for further exploration of its biological mechanisms.
[0052] The method for predicting the risk of atherosclerotic events provided by this invention determines the metabolic inflammatory liver index and the hepatic vascular protective biomarker index based on baseline clinical data. A multivariate Cox proportional hazards prediction model is established based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index, and traditional atherosclerotic risk variables. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the corresponding risk of atherosclerotic events is output. This method can significantly improve the ability to discriminate medium- and long-term risks, providing a new reference with potential clinical translational value for the risk assessment of atherosclerotic cardiovascular diseases.
[0053] The atherosclerotic event risk prediction device provided by the present invention will be described below. The atherosclerotic event risk prediction device described below can be referred to in correspondence with the atherosclerotic event risk prediction method described above.
[0054] Figure 5 A schematic diagram of the structure of an atherosclerotic event risk prediction device provided by the present invention is shown below. Figure 5 The device includes an acquisition module 51, an establishment module 52, and a prediction module 53, wherein: The acquisition module is used to acquire baseline clinical data and determine metabolic inflammatory liver index and hepatic vascular protective biomarker index based on the baseline clinical data; the baseline clinical data includes traditional atherosclerosis risk variables; A module is established to build a multivariate Cox proportional hazards model based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index and traditional atherosclerosis risk variables. The prediction module is used to output the risk of occurrence of corresponding atherosclerotic events based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model.
[0055] Since the apparatus of this embodiment is based on the same principle as the method of the above embodiment, more detailed explanations will not be repeated here.
[0056] It should be noted that, in the embodiments of the present invention, the relevant functional modules can be implemented by a hardware processor.
[0057] The atherosclerotic event risk prediction device provided by this invention determines the metabolic inflammatory liver index and the hepatic vascular protective biomarker index based on baseline clinical data. Based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index, and traditional atherosclerotic risk variables, a multivariate Cox proportional hazards prediction model is established. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the device outputs the corresponding risk of atherosclerotic events. This significantly improves the ability to discriminate medium- and long-term risks, providing a new and potentially clinically valuable reference for risk assessment of atherosclerotic cardiovascular diseases.
[0058] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include: a processor 61, a communication interface 62, a memory 63, and a communication bus 64, wherein the processor 61, the communication interface 62, and the memory 63 communicate with each other via the communication bus 64. The processor 61 can call logical instructions in the memory 63 to execute an atherosclerotic event risk prediction method, which includes: Baseline clinical data was obtained, and metabolic inflammatory liver indices and hepatic vascular protective biomarker indices were determined based on the baseline clinical data; the baseline clinical data included traditional atherosclerosis risk variables; Based on the aforementioned metabolic inflammatory liver index, hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, a multivariate Cox proportional hazards model was established. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the corresponding risk of atherosclerotic events is output.
[0059] Furthermore, the logical instructions in the aforementioned memory 63 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0060] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing a method for predicting the risk of atherosclerotic events, the method comprising: Baseline clinical data was obtained, and metabolic inflammatory liver indices and hepatic vascular protective biomarker indices were determined based on the baseline clinical data; the baseline clinical data included traditional atherosclerosis risk variables; Based on the aforementioned metabolic inflammatory liver index, hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, a multivariate Cox proportional hazards model was established. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the corresponding risk of atherosclerotic events is output.
[0061] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform a method for predicting the risk of atherosclerotic events, the method comprising: Baseline clinical data was obtained, and metabolic inflammatory liver indices and hepatic vascular protective biomarker indices were determined based on the baseline clinical data; the baseline clinical data included traditional atherosclerosis risk variables; Based on the aforementioned metabolic inflammatory liver index, hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, a multivariate Cox proportional hazards model was established. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the corresponding risk of atherosclerotic events is output.
[0062] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0063] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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.
Claims
1. A method for predicting the risk of atherosclerotic events, characterized in that, include: Baseline clinical data was obtained, and metabolic inflammatory liver indices and hepatic vascular protective biomarker indices were determined based on the baseline clinical data; the baseline clinical data included traditional atherosclerosis risk variables; Based on the aforementioned metabolic inflammatory liver index, hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, a multivariate Cox proportional hazards prediction model was established. Based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model, the corresponding risk of atherosclerotic events is output.
2. The method for predicting the risk of atherosclerotic events according to claim 1, characterized in that, The process involves establishing a multivariate Cox proportional hazards prediction model based on the metabolic inflammatory liver index, hepatic vascular protective biomarker index, and traditional atherosclerosis risk variables, including: A univariate Cox regression strategy was used to select candidate variables related to atherosclerotic events from traditional atherosclerosis risk variables; The candidate variables were screened using the LASSO-Cox regression strategy, and redundant variables were removed to obtain the variables used in the model. The model was trained based on the selected input variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index. The sensitivity and goodness of fit of the model were evaluated using time-dependent ROC curves, C-index, and AIC. When the sensitivity and evaluation values met the conditions, a multivariate Cox proportional hazards prediction model was obtained.
3. The method for predicting the risk of atherosclerotic events according to claim 2, characterized in that, The model is trained based on the selected input variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index. The sensitivity and goodness-of-fit of the model are evaluated using time-dependent ROC curves, C-index, and AIC. When the sensitivity and evaluation values meet the conditions, a multivariate Cox proportional hazards prediction model is obtained, including: The input variables, including the selected variables, metabolic inflammatory liver index, and hepatic vascular protective biomarker index, were used as input variables. Atherosclerosis-related events were used as endpoint events, and follow-up time was used as the time variable. A multivariate Cox proportional hazards regression method was used to fit the model, and the regression coefficients of each variable were obtained. An atherosclerosis risk prediction model was constructed, and the sensitivity and goodness of fit of the model were evaluated to obtain a multivariate Cox proportional hazards prediction model that can quantify the individual's risk of developing the disease.
4. The method for predicting the risk of atherosclerotic events according to claim 1, characterized in that, The metabolic inflammatory liver index was determined by body mass index (BMI), lymphocyte-to-neutrophil ratio, and C-reactive protein; the hepatic vascular protective biomarker index was determined by total bilirubin, albumin, alkaline phosphatase, and aspartate aminotransferase.
5. The method for predicting the risk of atherosclerotic events according to claim 2, characterized in that, The method employs a univariate Cox regression strategy to select candidate variables related to atherosclerotic events from traditional atherosclerosis risk variables, including: Univariate Cox proportional hazards regression models were constructed for each traditional atherosclerosis risk variable; Using atherosclerotic events as the endpoint and follow-up time as the time scale, we fitted and calculated the significance values of each variable one by one, and selected the variables with significance values less than the preset values as candidate variables.
6. The method for predicting the risk of atherosclerotic events according to claim 2, characterized in that, The LASSO-Cox regression strategy is used to screen the candidate variables, remove redundant variables, and obtain the variables to be included in the model, including: The LASSO-Cox regression model was used to screen candidate variables by penalized regression, and the optimal penalty coefficient λ was determined by ten-fold cross-validation. The variable coefficients are compressed based on the optimal penalty coefficient λ. Redundant variables whose coefficients are compressed to zero are removed, and variables with non-zero coefficients are retained as input variables.
7. The method for predicting the risk of atherosclerotic events according to claim 1, characterized in that, The atherosclerotic events include one or more of myocardial infarction, ischemic stroke, and revascularization surgery.
8. A device for predicting the risk of atherosclerotic events, characterized in that, include: The acquisition module is used to acquire baseline clinical data and determine metabolic inflammatory liver index and hepatic vascular protective biomarker index based on the baseline clinical data; the baseline clinical data includes traditional atherosclerosis risk variables; A module is established to build a multivariate Cox proportional hazards model based on the metabolic inflammatory liver index, the hepatic vascular protective biomarker index and traditional atherosclerosis risk variables. The prediction module is used to output the risk of occurrence of corresponding atherosclerotic events based on the clinical data to be predicted and the multivariate Cox proportional hazards prediction model.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the atherosclerotic event risk prediction method as described in any one of claims 1-7.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the atherosclerotic event risk prediction method as described in any one of claims 1-7.