Association Between Dietary Vitamin Intake and Severe Abdominal Aortic Calcification in U.S. Adults: A SHAP-Based Machine Learning Analysis of NHANES 2013–2014

Association Between Dietary Vitamin Intake and Severe Abdominal Aortic Calcification in U.S. Adults: A SHAP-Based Machine Learning Analysis of NHANES 2013–2014 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Association Between Dietary Vitamin Intake and Severe Abdominal Aortic Calcification in U.S. Adults: A SHAP-Based Machine Learning Analysis of NHANES 2013–2014 Dingwei Lei, Zhuo Li, Shihan Guo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6687344/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Severe abdominal aortic calcification (SAAC) significantly impacts families and society. Although vitamin intake is closely linked to SAAC, large-scale model-based dietary studies are scarce. This study compares machine learning models to analyze this association and support dietary strategies for SAAC prevention. Methods: There are 10 ways to build ML models. The best model for further analysis was selected based on accuracy, area under the subject operating characteristic curve (AUC), precision, recall, and F1 score. Shapley Additive Explanations (SHAP) method was used to explain the contribution of variables to ML models. Results: Logistic Regression (LR) exhibited the best performance in exploring the association between dietary vitamins and SAAC, with an AUC of 0.851. The SHAP values indicate that among dietary vitamins, vitamin A has the greatest contribution to the machine learning (ML) model. Age is the most important feature among all characteristics, while folate has the least impact on the ML model. Conclusion: LR algorithm performed best, vitamin A was the most significant factor, folic acid correlation was weak.The model is helpful for early screening and intervention of SAAC and improves patient prognosis. SAAC Dietary vitamin Machine learning NHANES AUC Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 1 Introduction Severe abdominal aortic calcification (AAC) serves as a definitive marker of subclinical atherosclerosis and a critical predictor of cardiovascular disease (CVD) ( 1 ) .Cardiovascular disease is the leading cause of non-communicable diseases worldwide, imposing a significant burden on families and society( 2 ). In the United States, the incidence of AAC increases with age, and its prevalence continues to rise annually ( 3 ). Although studies suggest that the exact mechanism of vascular calcification likely involves elements of bone metabolism, our understanding of the mechanisms underlying AAC remains incomplete, and effective prevention and treatment strategies are still lacking( 4 – 6 ). Therefore, early detection and a comprehensive understanding of the risk factors associated with AAC are crucial for effective cardiovascular risk assessment and management. Diet plays a crucial role in vascular health, and previous studies have demonstrated that dietary factors, such as the intake of vitamins, are associated with the incidence of abdominal aortic calcification( 7 – 10 ). This association may be attributed to the vitamins' ability to scavenge free radicals and inhibit oxidation. However, existing studies have primarily focused on the individual effects of specific dietary vitamin intakes, and research on the combined effects of dietary and supplemental sources of vitamin intake remains insufficient. Given the potential interactions among different dietary vitamins, examining a single micronutrient in isolation fails to account for the synergistic effects of dietary vitamin intake on the risk of severe abdominal aortic calcification. Therefore, further investigation into the protective role of dietary vitamins in this complex pathology is of significant importance. Our study aims to utilize the National Health and Nutrition Examination Survey (NHANES) to identify potential associations between dietary vitamin intake and severe abdominal aortic calcification through machine learning (ML) methods. Unlike traditional statistical approaches, ML techniques can manage large and complex datasets while uncovering implicit relationships among various health-related features, thus enabling more accurate predictions of disease risk( 11 , 12 ). These early prediction models will aid in controlling the extent of abdominal aortic calcification and enhance public awareness of prevention, thereby making a significant contribution to reducing the socioeconomic burden and adverse public health events associated with this irreversible and severe vascular disease. 2 Methods 2.1 Study population NHANES is a complex, multistage sampling program conducted by the Centers for Disease Control and Prevention (CDC) that serves as the data source for the analysis in this study. The survey received approval from the Institutional Review Board (IRB), and written consent was obtained from all participants. Participants from the 2013–2014 NHANES were considered candidates for this study. Cases with missing data for either independent or dependent variables were excluded, and the remaining participants were included for further analysis. 2.2 Dietary vitamin intake The intake data of 11 dietary vitamins were obtained from NHANES. Participants underwent two 24-hour dietary recall interviews at mobile examination centers, with an interval of 3 to 10 days between the two interviews. The average daily intake of dietary antioxidants was subsequently calculated. 2.3 AAC Abdominal aortic calcification (AAC) was defined using the Kauppila scoring system, which employs dual-energy X-ray absorptiometry. The assessment divides the abdominal aortic wall into four segments, with each segment scored from 0 to 6 based on the level of calcium deposition. This results in a cumulative AAC score ranging from 0 to 24. Typically, an AAC score greater than 6 is considered indicative of severe AAC( 13 , 14 ). 2.4 Covariates Demographic data were obtained using standardized surveys that covered gender, race/ethnicity, educational background, smoking habits, alcohol consumption, poverty income ratio (PIR), body mass index (BMI), and blood pressure (BP). Participants were categorized into the drinking group and the non-drinking group based on their response to the question, "Do you drink 4/5 cups or more of alcohol per day?" Smoking status was assessed as never smoker (fewer than 100 cigarettes), former smoker (not currently smoking but having smoked 100 or more cigarettes), or current smoker (having smoked 100 or more cigarettes and currently smoking every day or on some days). The definition of hypertension was based on the question, "Has your doctor ever told you that you have hypertension?" The criteria for diagnosing diabetes include any of the following: ( 1 ) HbA1c ≥ 6.5%; ( 2 ) fasting plasma glucose (FPG) ≥ 7.0 mmol/L; ( 3 ) random plasma glucose (RPG) ≥ 11.1 mmol/L or oral glucose tolerance test (OGTT) ≥ 11.1 mmol/L; ( 4 ) a doctor’s diagnosis of diabetes; or ( 5 ) currently taking anti-diabetic medications or insulin( 15 ). 2.5 Correlation Analysis and Feature Selection Multicollinearity refers to the correlation between two or more variables. In practice, when variables are correlated, interpreting the impact of an independent variable on the dependent variable becomes challenging. Specifically, multiple correlated independent variables jointly influence the dependent variable. Consequently, the presence of multicollinearity reduces the reliability of the estimated coefficients and diminishes the stability and performance of the model. To detect multicollinearity among the included features, we generally use correlation coefficients and variance inflation factors. We consider that multicollinearity exists among independent variables when the correlation coefficient exceeds 0.9 or the variance inflation factor exceeds 10. Since the binary covariates and vitamin concentrations in the statistical model are non-normally distributed, Spearman correlation analysis is employed to detect associations among all independent variables. Additionally, due to the excessive number of independent variables identified by the model, we employed the Boruta algorithm and the Least Absolute Shrinkage and Selection Operator (LASSO) for feature selection. 2.6 Machine learning model building To establish a predictive model for severe abdominal aortic calcification and assess its predictive capability, the data were randomly divided into a training set and a test set in a 7:3 ratio. To address the issue of class imbalance in the initial dataset, SMOTE was applied to the training set to enhance the performance of the machine learning (ML) models. The algorithms employed in this study included Neural Networks (NN), Random Forest (RF), Extreme Gradient Boosting (XGB), Gradient Boosting Machine (GBM), Logistic Regression (LR), Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Light Gradient Boosting Machine (LGB), Adaptive Boosting (Adaboost), and Categorical Boosting (CatBoost), totaling ten ML algorithms to predict the risk of severe abdominal aortic calcification. The constructed models were subsequently applied to the test set. Additionally, the project team is developing a new technology for Xiaomi. 2.7 Model evaluation and interpretation For the validation set, the performance of the machine learning model was evaluated using several metrics: accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve (AUC). The model exhibiting the best discriminative power was identified. Subsequently, calibration curves and decision curves were plotted to further validate the model's predictive ability and practical application value. The model was interpreted using Shapley Additive Explanations (SHAP). All analyses and modeling in this study were conducted using the statistical software R version 4.4.2 and RStudio. All tests were two-sided, and a p-value of less than 0.05 was considered statistically significant. 3 Results 3.1 Baseline characteristics of study populations Figure 1 illustrates the screening process for the study subjects, including a total of 1,070 adults with complete data. Among these, 993 individuals did not have severe abdominal aortic calcification (SAAC), while 137 did. Table 1 presents the baseline characteristics of the study population. Patients with severe SAAC generally exhibited lower dietary vitamin content compared to those without severe SAAC. Furthermore, comparisons of baseline characteristics between the two groups revealed statistically significant differences in age, ethnicity, BMI, smoking status, education level, hypertension, and diabetes (P < 0.05). Table 1 Baseline characteristics of the study population Characteristic Non-Sever AAC N = 933 (89%) Sever AAC N = 137 (11%) P -Value Age 57.00 (11.40) 70.51 (9.42) < 0.001 Sex 0.388 Female 497.00(53.27%) 67.00(48.91%) Male 436.00 (46.73%) 70.00 (51.09%) Race 0.01 Mexican American 121.00 (12.97%) 15.00 (10.95%) Other Hispanic 85 .00(9.11%) 6 .00(4.38%) Non-Hispanic White 440.00 (47.16%) 86 .00(62.77%) Non-Hispanic Black 186 .00(19.94%) 22.00 (16.06%) Other 101.00 (10.83%) 8.00 (5.84%) BMI 28.69 (5.65) 26.91 (4.46) < 0.001 Alcohol 0.487 No 769.00 (84.33%) 109.00 (78.54%) Yes 164.00 (15.67%) 28.00 (21.46%) Smoke =High school 751.00 (86.99%) 98.00 (74.63%) <High school 182.00 (13.01%) 39.00 (25.37%) PIR 2.79 (1.65) 2.65 (1.59) 0.324 Hypertension < 0.001 No 518.00 (58.32%) 43.00 (33.43%) Yes 415.00 (41.68%) 94.00 (66.57%) Diabetes < 0.001 No 745.00 (84.31%) 89.00 (69.74%) Yes 188.00 (15.69%) 48.00 (30.26%) folicacid 233.89 (137.58) 183.50 (80.74) < 0.001 Vitamin A 629.44 (454.86) 535.26 (328.23) 0.02 VitaminB1 1.59 (0.74) 1.37 (0.58) 0.001 VitaminB2 2.10 (1.05) 1.85 (0.79) 0.006 VitaminB6 2.11 (1.17) 1.75 (0.83) 0.001 VitaminB12 4.84 (3.97) 3.94 (2.75) 0.01 Vitamin C 80.59 (73.74) 77.28 (81.36) 0.628 Vitamin D 5.08 (5.18) 4.16 (3.31) 0.046 Vitamin E 0.79 (3.16) 0.37 (1.49) 0.122 Vitamin K 130.73 (149.18) 91.43 (75.06) 0.003 Niacin 25.50 (12.97) 21.79 (9.20) 0.001 3.2 Correlation Analysis and Feature Selection We explored the correlations between dietary vitamins and other independent variables using Spearman correlation analysis. The results of these correlations, presented in Fig. 2 , show the correlation coefficients (r) between dietary vitamins and participants' baseline characteristics. Most vitamins exhibited varying degrees of correlation, with the most significant correlation observed between niacin and vitamin B6 (r = 0.81). This was followed by strong correlations between niacin and vitamin B1 (r = 0.71) and vitamin B2 (r = 0.68). In this study, we defined a correlation coefficient greater than 0.9 as indicative of multicollinearity. Furthermore, the variance inflation factor (VIF) values of the included covariates were all below 10 (Fig. 3 ), indicating the absence of multicollinearity among the variables. We divided the data into a training group consisting of 749 cases and a testing group comprising 321 cases, maintaining a 7:3 ratio. Statistical analysis revealed no significant differences between the two groups (Table 2 ). LASSO regression, a shrinkage estimation method, achieves variable selection and complexity adjustment by formulating an optimization objective function that includes a penalty term. In this study, LASSO regression was employed to identify characteristic factors such as folic acid, diabetes, smoking, vitamin A, age, and BMI (Fig. 4 A, B). The Boruta algorithm, an extension of the RF algorithm, accurately estimates the importance of each feature, thereby identifying the true set of features. We utilized the Boruta algorithm to identify 14 key factors, including age, BMI, diabetes, hypertension, folic acid, vitamin A, vitamin B, vitamin C, vitamin D, and niacin (Fig. 4 C). Through comparative analysis of the screening results from LASSO regression and the Boruta algorithm, we determined the subset of common feature variables selected by both methods, ultimately narrowing down to five feature variables: diabetes, folic acid, vitamin A, age, and BMI (Fig. 4 D). Table 2 test train p.overall N = 321 N = 749 SEQN 78697 (2883) 78704 (2958) 0.969 Result: 0.281 0 274 (85.4%) 659 (88.0%) 1 47 (14.6%) 90 (12.0%) Sex: 0.400 1 176 (54.8%) 388 (51.8%) 2 145 (45.2%) 361 (48.2%) Race: 0.439 1 43 (13.4%) 93 (12.4%) 2 20 (6.23%) 71 (9.48%) 3 165 (51.4%) 361 (48.2%) 4 59 (18.4%) 149 (19.9%) 5 34 (10.6%) 75 (10.0%) Educational: 0.717 1 69 (21.5%) 152 (20.3%) 2 252 (78.5%) 597 (79.7%) Alcohol: 0.289 0 270 (84.1%) 608 (81.2%) 1 51 (15.9%) 141 (18.8%) Smoke: 0.351 1 147 (45.8%) 364 (48.6%) 2 120 (37.4%) 224 (29.9%) 3 54 (16.8%) 161 (21.5%) Hypertension: 0.979 0 169 (52.6%) 392 (52.3%) 1 152 (47.4%) 357 (47.7%) Diabetes: 0.394 0 256 (79.8%) 578 (77.2%) 1 65 (20.2%) 171 (22.8%) Age 58.8 (12.0) 58.7 (12.1) 0.887 PIR 2.82 (1.63) 2.76 (1.65) 0.595 BMI 28.1 (5.33) 28.6 (5.63) 0.165 folicacid 227 (124) 228 (136) 0.902 VitaminA 614 (456) 619 (436) 0.880 VitaminB1 1.57 (0.77) 1.56 (0.70) 0.985 VitaminB2 2.10 (1.14) 2.06 (0.97) 0.630 VitaminB6 2.17 (1.29) 2.02 (1.06) 0.058 VitaminB12 4.96 (4.14) 4.62 (3.71) 0.200 Vitamin C 89.4 (79.5) 76.2 (72.3) 0.051 Vitamin D 5.30 (5.27) 4.81 (4.86) 0.153 Vitamin E 0.75 (2.99) 0.73 (3.01) 0.933 Vitamin K 119 (142) 129 (143) 0.309 Niacin 25.9 (14.7) 24.7 (11.6) 0.178 Predictor screening results. (A) LASSO regression model screening variable trajectories; (B) Factor screening based on the LASSO regression model, with the left dashed line indicating the best lambda value for the evaluation metrics (lambda. min) and the right dashed line indicating the lambda value for the model where the evaluation metrics are in the range of the best value by one standard error (lambda.1se); (C) Boruta; (D) common predictors between Boruta and LASSO. 3.3 Evaluation of ML Models Based on the training data, we plotted the ROC curve (Fig. 5 ), which revealed that the Logistic model exhibited the best predictive performance, achieving an AUC of 0.851, indicative of high prediction accuracy. By integrating metrics such as accuracy, precision, recall, and F1 score, we conducted a comprehensive evaluation of the model's performance, concluding that the Logistic model outperformed the others (Table 3 ). Consequently, the Logistic model was selected for further analysis. Table 3 ML performance index Model AUC Accuracy Sensitivity Specificity Precision F1 Logistic 0.851(0.792–0.909) 0.787 0.78 0.789 0.352 0.485 SVM 0.833(0.772–0.895) 0.234 0.195 0.24 0.036 0.061 GBM 0.807(0.738–0.876) 0.716 0.805 0.703 0.284 0.42 KNN 0.806(0.744–0.868) 0.566 0.976 0.505 0.225 0.365 RF 0.792(0.723–0.862) 0.713 0.805 0.699 0.282 0.418 XGB 0.793(0.713–0.873) 0.803 0.732 0.814 0.366 0.488 KNN 0.806(0.744–0.868) 0.566 0.976 0.505 0.225 0.365 Adaboost 0.743(0.678–0.807) 0.678 0.829 0.656 0.262 0.398 LGB 0.666(0.554–0.777) 0.719 0.659 0.728 0.262 0.375 CatBoost 0.839(0.775–0.902) 0.691 0.951 0.652 0.287 0.441 Since the discriminative performance had minimal impact on the comparative predictive performance of the models, this study proceeded to evaluate the calibration degree of each model through calibration curves. The calibration curve of the Logistic model in the validation cohort exhibited good consistency, closely following the ideal diagonal line (Fig. 6 ), which indicates high predictive accuracy. Furthermore, the Decision Curve Analysis (DCA) in the validation cohort demonstrated that across a wide range of threshold probabilities, the net benefit of this model consistently exceeded that of the two extreme strategies (treat-all and treat-none), suggesting its potential clinical utility (Fig. 7 ). In summary, our evaluation results indicate that our machine learning model can effectively identify and benefit patients with severe abdominal aortic calcification. Finally, we plotted the SHAP diagram (Fig. 8 ) to illustrate the impact of various features on the model's output. Age had the most significant influence, with higher values contributing positively and lower values contributing negatively; notably, a considerable proportion of individuals had a negative impact on the prediction. Following this, diabetes also exerted a negative influence, predominantly among individuals with lower values. Vitamin A ranked fourth in its effect on the model's output, where higher values primarily contributed negatively and lower values contributed positively. Folic acid had the least impact on the model's predictions, with lower values mainly contributing positively. The results indicated that the two most significant potential key heavy metals associated with SAAC risk are Vitamin A and folic acid. Construction of nomograms This study integrated five significant predictive variables—Diabetes, folic acid, Vitamin A, Age, and BMI—to intuitively assess the risk of severe abdominal aortic calcification. Figure 9 presents the nomogram, which offers a rapid and easily interpretable risk assessment. 4 Discussion Severe abdominal aortic calcification poses a significant threat to human health. In previous studies, we identified a correlation between various dietary vitamins and abdominal aortic calcification, with the majority acting as protective factors against its occurrence. To date, we have not discovered any predictive models for severe abdominal aortic calcification. Dietary vitamin intake, as a modifiable lifestyle factor, exhibits not only long-term stability but also clear value for intervention guidance. Particularly at the public health level, prediction models based on nutritional behaviors possess stronger generalizability and preventive potential. Therefore, this study aims to address this gap by developing a predictive model for severe abdominal aortic calcification based on dietary vitamins, providing a new perspective for early risk assessment and potential dietary interventions. This study explores the complex relationship between dietary vitamins and severe abdominal aortic calcification (SAAC). Among the ten machine learning algorithms, the Logistic model excelled in identifying potential risks of SAAC (AUC = 0.851). SHAP values indicated that, among dietary vitamins, vitamin A contributed the most to the machine learning model. Age was the most significant feature among all characteristics, while folic acid had the least impact on the machine learning model. This model serves as a valuable auxiliary tool for clinical decision-making. This study indicates that among dietary vitamins, Vitamin A exhibits the strongest correlation with SAAC. In a study focused on identifying dietary components associated with abdominal aortic calcification, data on 35 macronutrients and micronutrients were analyzed, revealing that Vitamin A concentration is inversely correlated with abdominal aortic calcification( 16 ). Currently, the literature on the association between vitamin A intake and aortic arch calcification (AAC) remains limited. Due to its antioxidant properties, vitamin A supplementation may be linked to protective effects at the arterial level( 17 ). However, considering the potential toxicity of high doses of vitamin A, further research is essential to investigate its application as a protective measure against AAC. In this study, we found that age is a significant predictor of SAAC, which aligns with common understanding. The atherosclerotic process begins in childhood and may progress to the formation of advanced atherosclerotic plaques, with some lesions undergoing calcification. Multiple studies have explored the relationship between aortic calcification and age, all concluding that abdominal aortic calcification is positively correlated with age( 18 – 22 ). Current literature searches have not identified any studies demonstrating a negative or zero correlation with age, suggesting that age is a significant determinant of the presence and severity of abdominal aortic calcification. However, diabetes mellitus is regarded as a critical factor in this study, as it is associated with arterial calcification, including both intimal arterial calcification and atherosclerotic intimal calcification. Multiple studies support the relationship between aortic calcification and diabetes ( 18 , 22 ). In contrast, Matsushita et al.( 23 ) did not find a correlation between diabetes and the incidence or severity of calcification when evaluating a male cohort; however, this may be attributed to the small sample size of the study. Moreover, the association between BMI and SAAC should not be overlooked. Recent studies suggest that lean body mass may be a risk factor for AAC( 24 , 25 ), and their findings align with those of the current study. Additionally, visceral fat accumulation has been shown to be closely associated with dyslipidemia, insulin resistance, diabetes, and hypertension, all of which increase the risk of cardiovascular disease( 26 – 28 ). This discrepancy may be due to BMI's limitations in accurately reflecting fat distribution( 27 ). Folic acid contributes the least to the machine learning (ML) model. Previous studies have demonstrated that folic acid serves as a protective factor against severe aortic arch calcification (AAC) formation, with the benefits being most pronounced in individuals who have adequate folic acid intake, particularly those with a daily intake exceeding 280.126 mg ( 29 ). A potential contributing factor to this association is the regulation of homocysteine levels. Research indicates that folic acid supplementation effectively reduces hyperhomocysteinemia, which is a recognized trigger for vascular dysfunction and an increased risk of cardiovascular disease ( 30 , 31 ). Additionally, oxidative stress can lead to vascular calcification, and folic acid deficiency exacerbates oxidative stress and various aspects of metabolic syndrome, both of which are associated with an elevated risk of diabetes and cardiovascular diseases( 32 , 33 ). Although there has been considerable speculation regarding the role of folic acid in arterial calcification, the specific mechanisms through which its dietary intake influences the occurrence of severe AAC still require in-depth exploration. However, this study has several limitations. Firstly, while the LR model demonstrated strong predictive performance, its generalizability remains uncertain due to the lack of external validation. Secondly, the manually selected features may introduce slight bias, suggesting that advanced algorithms could save time and costs for researchers by utilizing automatic feature selection techniques. Thirdly, dietary vitamin intake is often associated with geographical location; however, this variable was not included in the model. These factors should be considered in future research. Finally, as a cross-sectional study, this research cannot verify causal associations due to the absence of time series data, and the results should be interpreted with caution. 5 Conclusion This study constructed a machine learning (ML) model to explore the association between the risk of SAAC and dietary vitamins, while also incorporating demographic data. The logistic regression (LR) algorithm demonstrated superior performance compared to the other nine algorithms assessed. Among the eleven vitamins examined, vitamin A emerged as the most significant factor in the ML model, whereas folate exhibited a weaker association with SAAC than vitamin A. Among all independent variables, age was identified as the most crucial factor in this model. Furthermore, diabetic patients exhibited a higher risk of developing SAAC. In contrast to obese individuals, lean body mass may pose a risk factor for AAC. Future research should focus on developing more advanced algorithms to validate these findings and adjust relevant parameters to enhance the model's accuracy. Abbreviations Severe abdominal aortic calcification SAAC Mchine learning ML National Health and Nutrition Examination Survey NHANES Least Absolute Shrinkage and Selection Operator LASSO Synthetic Minority Over-sampling Technique SMOTE Neural Networks NN Random Forests RF Extreme Gradient Boosting XGB Gradient Boosting Machines GBM Logistic Regression LR Support Vector Machine SVM k-Nearest Neighbors KNN Light Gradient Boosting Machine LGB Adaptive Boosting Adaboost the area under the Receiver Operating Characteristic curve AUC The Shapley Additive Explanations SHAP Crdiovascular disease CVD Te Centers for Disease Control and Prevention CDC The Institutional Review Board IRB Poverty income ratio PIR Body mass index BMI Blood pressure BP The variance inflation factor VIF Decision Curve Analysis DCA Declarations Ethics approval and consent to participate The research involving human subjects received approval from the Institutional Review Board of the National Center for Health Statistics. The study was conducted in accordance with local legislation and institutional requirements, and adhered to the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants for their involvement in this research. Consent for publication All authors of this study have fully discussed and unanimously agreed to submit this manuscript for publication. Data availability statement The complete list of datasets supporting the findings of this study is available at the following URL:https://www.cdc.gov/nchs/nhanes/。 Conflict of interest The author declares that this research was conducted without any commercial or financial relationships that could be perceived as a potential conflict of interest. Funding The author(s) declare that no financial support was received for the research and/or publication of this article. Author contributions DW: Data Management, Writing-Original Draft. ZL: Software, validation, writing-review and edit. SG: Funding, supervision, writing-reviewing and editing. References Allam AHA, Thompson RC, Eskander MA, Mandour Ali MA, Sadek A, Rowan CJ, Sutherland ML, Sutherland JD, Frohlich B, Michalik DE, Finch CE, Narula J, Thomas GS, Samuel Wann L. Is coronary calcium scoring too late? Total body arterial calcium burden in patients without known CAD and normal MPI. J nuclear cardiology: official publication Am Soc Nuclear Cardiol. 2018;25:1990–8. 10.1007/s12350-017-0925-9 . Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, Barengo NC, Beaton AZ, Benjamin EJ, Benziger CP, Bonny A, Brauer M, Brodmann M, Cahill TJ, Carapetis J, Catapano AL, Chugh SS, Cooper LT, Coresh J, Criqui M, DeCleene N, Eagle KA, Emmons-Bell S, Feigin VL, Fernández-Solà J, Fowkes G, Gakidou E, Grundy SM, He FJ, Howard G, Hu F, Inker L, Karthikeyan G, Kassebaum N, Koroshetz W, Lavie C, Lloyd-Jones D, Lu HS, Mirijello A, Temesgen AM, Mokdad A, Moran AE, Muntner P, Narula J, Neal B, Ntsekhe M, de Moraes G, Otto C, Owolabi M, Pratt M, Rajagopalan S, Reitsma M, Ribeiro ALP, Rigotti N, Rodgers A, Sable C, Shakil S, Sliwa-Hahnle K, Stark B, Sundström J, Timpel P, Tleyjeh IM, Valgimigli M, Vos T, Whelton PK, Yacoub M, Zuhlke L, Murray C, Fuster V. Global Burden of Cardiovascular Diseases and Risk Factors, 1990–2019: Update From the GBD 2019 Study. J Am Coll Cardiol. 2020;76:2982–3021. 10.1016/j.jacc.2020.11.010 . Wong ND, Lopez VA, Allison M, Detrano RC, Blumenthal RS, Folsom AR, Ouyang P, Criqui MH. Abdominal aortic calcium and multi-site atherosclerosis: the Multiethnic Study of Atherosclerosis. Atherosclerosis .(2011)214:436 – 41. 10.1016/j.atherosclerosis.2010.09.011 Pan W, Jie W, Huang H. Vascular calcification: Molecular mechanisms and therapeutic interventions. MedComm (2023)4:e200. 10.1002/mco2.200 Doherty TM, Fitzpatrick LA, Inoue D, Qiao JH, Fishbein MC, Detrano RC, Shah PK, Rajavashisth TB. Molecular, endocrine, and genetic mechanisms of arterial calcification. Endocr Rev. 2004;25:629–72. 10.1210/er.2003-0015 . McCullough PA, Soman S. Cardiovascular calcification in patients with chronic renal failure: are we on target with this risk factor? Kidney Int Supplement. 2004;S18–24. 10.1111/j.1523-1755.2004.09008.x . Li H, Li R, Gong C, Wu Z, Jia Q. The relationship between dietary vitamin B1 intake and severe abdominal aortic calcification among the general population in the United States. Front Cardiovasc Med. 2024;11:1415151. 10.3389/fcvm.2024.1415151 . El Maghraoui A, Hamza T, Sadni S, El Maataoui A, Majjad A, Rezqi A, Ouzzif Z, Mounach A. Vitamin D status and abdominal aortic calcification in postmenopausal women. J Bone Miner Metab. 2018;36:229–37. 10.1007/s00774-017-0832-9 . Jia J, Zhang J, He Q, Wang M, Liu Q, Wang T, Chen X, Wang W, Xu H. Association between dietary vitamin C and abdominal aortic calcification among the US adults. Nutr J. 2023;22:58. 10.1186/s12937-023-00889-y . Oyama S, Okamoto N, Koide S, Hayashi H, Nakai S, Takahashi K, Inaguma D, Hasegawa M, Toyama H, Sugiyama S, Yuzawa Y, Tsuboi N. Vitamin K2 supplementation and the progression of abdominal aortic calcification in dialysis patients. Fujita Med J. 2021;7:136–8. 10.20407/fmj.2020-020 . Rajula HSR, Verlato G, Manchia M, Antonucci N, Fanos V. Comparison of Conventional Statistical Methods with Machine Learning in Medicine: Diagnosis, Drug Development, and Treatment. Med (Kaunas Lithuania). 2020;56. 10.3390/medicina56090455 . Li X, Zhao Y, Zhang D, Kuang L, Huang H, Chen W, Fu X, Wu Y, Li T, Zhang J, Yuan L, Hu H, Liu Y, Zhang M, Hu F, Sun X, Hu D. Development of an interpretable machine learning model associated with heavy metals' exposure to identify coronary heart disease among US adults via SHAP: Findings of the US NHANES from 2003 to 2018. Chemosphere. 2023;311:137039. 10.1016/j.chemosphere.2022.137039 . Xie R, Liu X, Wu H, Liu M, Zhang Y. Associations between systemic immune-inflammation index and abdominal aortic calcification: Results of a nationwide survey. Nutr metabolism Cardiovasc diseases: NMCD. 2023;33:1437–43. 10.1016/j.numecd.2023.04.015 . Xiao Y, Quan Y. Associations between dietary fatty acids intake and abdominal aortic calcification: a national population-based study. Lipids Health Dis. 2024;23:73. 10.1186/s12944-024-02059-3 . Yin J, Zheng C, Lin X, Huang C, Hu Z, Lin S, Qu Y. The potential of the serum uric acid to high-density lipoprotein cholesterol ratio as a predictive biomarker of diabetes risk: a study based on NHANES 2005–2018. Front Endocrinol. 2024;15:1499417. 10.3389/fendo.2024.1499417 . Li W, Huang G, Tang N, Lu P, Jiang L, Lv J, Qin Y, Lin Y, Xu F, Lei D. Identification of dietary components in association with abdominal aortic calcification. Food Funct. 2023;14:8383–95. 10.1039/d3fo02920d . da Rocha RF, de Oliveira MR, Schonhofen P, De Bastiani MA, Schnorr CE, Klamt F, Dal Pizzol F, Moreira JC. Vitamin A supplementation for different periods alters rat vascular redox parameters. J Physiol Biochem. 2010;66:351–7. 10.1007/s13105-010-0041-7 . Allison MA, Criqui MH, Wright CM. Patterns and risk factors for systemic calcified atherosclerosis. Arteriosclerosis, thrombosis, and vascular biology .(2004)24:331-6. 10.1161/01.ATV.0000110786.02097.0c Kiel DP, Kauppila LI, Cupples LA, Hannan MT, O'Donnell CJ, Wilson PW. Bone loss and the progression of abdominal aortic calcification over a 25 year period: the Framingham Heart Study. Calcif Tissue Int. 2001;68:271–6. 10.1007/bf02390833 . Miwa Y, Tsushima M, Arima H, Kawano Y, Sasaguri T. Pulse pressure is an independent predictor for the progression of aortic wall calcification in patients with controlled hyperlipidemia. Hypertension (Dallas, Tex : 1979).(2004)43:536 – 40. 10.1161/01.HYP.0000117153.48029.d1 O'Donnell CJ, Chazaro I, Wilson PW, Fox C, Hannan MT, Kiel DP, Cupples LA. Evidence for heritability of abdominal aortic calcific deposits in the Framingham Heart Study. Circulation .(2002)106:337 – 41. 10.1161/01.cir.0000022663.26468.5b Reaven PD, Sacks J. Reduced coronary artery and abdominal aortic calcification in Hispanics with type 2 diabetes. Diabetes Care. 2004;27:1115–20. 10.2337/diacare.27.5.1115 . Matsushita M, Nishikimi N, Sakurai T, Nimura Y. Relationship between aortic calcification and atherosclerotic disease in patients with abdominal aortic aneurysm. Int Angiol. 2000;19:276–9. Yuan M, Hsu FC, Bowden DW, Xu J, Carrie Smith S, Wagenknecht LE, Comeau ME, Divers J, Register TC, Jeffrey Carr J, Langefeld CD, Freedman BI. Relationships between measures of adiposity with subclinical atherosclerosis in patients with type 2 diabetes. Obesity (Silver Spring, Md) .(2016)24:1810-8. 10.1002/oby.21540 Rodríguez AJ, Scott D, Khan B, Khan N, Hodge A, English DR, Giles GG, Ebeling PR. Low Relative Lean Mass is Associated with Increased Likelihood of Abdominal Aortic Calcification in Community-Dwelling Older Australians. Calcif Tissue Int. 2016;99:340–9. 10.1007/s00223-016-0157-z . Bray GA, Heisel WE, Afshin A, Jensen MD, Dietz WH, Long M, Kushner RF, Daniels SR, Wadden TA, Tsai AG, Hu FB, Jakicic JM, Ryan DH, Wolfe BM, Inge TH. The Science of Obesity Management: An Endocrine Society Scientific Statement. Endocr Rev. 2018;39:79–132. 10.1210/er.2017-00253 . Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, Vasan RS, Murabito JM, Meigs JB, Cupples LA, D'Agostino RB, Sr., O'Donnell CJ. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116:39–48. 10.1161/circulationaha.106.675355 . Britton KA, Massaro JM, Murabito JM, Kreger BE, Hoffmann U, Fox CS. Body fat distribution, incident cardiovascular disease, cancer, and all-cause mortality. J Am Coll Cardiol. 2013;62:921–5. 10.1016/j.jacc.2013.06.027 . Zhang K, Chen J, Chen B, Han Y, Cai T, Zhao J, Gu Z, Gao M, Hou Z, Yu X, Gu F, Gao Y, Hu R, Xie J, Liu T, Cui D, Li B. Association between dietary folate intake and severe abdominal aorta calcification in adults: A cross-sectional analysis of the national health and nutrition examination survey. Diabetes Vasc Dis Res. 2024;21:14791641241246555. 10.1177/14791641241246555 . Li T, Dong G, Kang Y, Zhang M, Sheng X, Wang Z, Liu Y, Kong N, Sun H. Increased homocysteine regulated by androgen activates autophagy by suppressing the mammalian target of rapamycin pathway in the granulosa cells of polycystic ovary syndrome mice. Bioengineered. 2022;13:10875–88. 10.1080/21655979.2022.2066608 . Sung ML, Wu CC, Chang HI, Yen CK, Chen HJ, Cheng JC, Chien S, Chen CN. Shear stress inhibits homocysteine-induced stromal cell-derived factor-1 expression in endothelial cells. Circul Res. 2009;105:755–63. 10.1161/circresaha.109.206524 . Yamada S, Taniguchi M, Tokumoto M, Toyonaga J, Fujisaki K, Suehiro T, Noguchi H, Iida M, Tsuruya K, Kitazono T. The antioxidant tempol ameliorates arterial medial calcification in uremic rats: important role of oxidative stress in the pathogenesis of vascular calcification in chronic kidney disease. J bone mineral research: official J Am Soc Bone Mineral Res. 2012;27:474–85. 10.1002/jbmr.539 . Pravenec M, Kozich V, Krijt J, Sokolová J, Zídek V, Landa V, Simáková M, Mlejnek P, Silhavy J, Oliyarnyk O, Kazdová L, Kurtz TW. Folate deficiency is associated with oxidative stress, increased blood pressure, and insulin resistance in spontaneously hypertensive rats. Am J Hypertens. 2013;26:135–40. 10.1093/ajh/hps015 . Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6687344","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":496778355,"identity":"a7101dcb-3607-4093-b21c-fd57d3ee0f7e","order_by":0,"name":"Dingwei Lei","email":"","orcid":"","institution":"Chengdu University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Dingwei","middleName":"","lastName":"Lei","suffix":""},{"id":496778356,"identity":"85781f2f-bb3c-4135-92c3-0be4da84073e","order_by":1,"name":"Zhuo Li","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Zhuo","middleName":"","lastName":"Li","suffix":""},{"id":496778358,"identity":"509620bd-4862-44cc-95dc-b8ca3a6d851c","order_by":2,"name":"Shihan Guo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYBACNvbmAwckKmyY7c83JD5IqKghrIWP51jiA4szaewMNw48Nnhw5hhhLXISOcYGlW2H+RkOJD6TfNjCTITDJNLSJG6cOSzN2HA4rSKxgY2Bv707Ab8WnsfHJGdUpBszM7el3UjcIcMgcebsBvxa2NPSpCXOWCezMZwBajnDxmAgkUtAC0OOmfTfNub6Hob8bwWJbcxEaOEAel+yzZlZgiEhjYE4LaBAljiTxmwgcSBZIuHMMR6CfpFvh0alAX9D4scfFTVy/O29+LVgAB7SlI+CUTAKRsEowAoAyahNO0/jVXUAAAAASUVORK5CYII=","orcid":"","institution":"Hospital of Chengdu University of Traditional Chinese Medicine","correspondingAuthor":true,"prefix":"","firstName":"Shihan","middleName":"","lastName":"Guo","suffix":""}],"badges":[],"createdAt":"2025-05-17 13:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6687344/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6687344/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88642936,"identity":"de3be600-6856-4189-bd37-f56c195b3706","added_by":"auto","created_at":"2025-08-08 16:12:46","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1108285,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of participant selection.\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/15201d96e76184a9a553737d.png"},{"id":88642931,"identity":"83ed2b5f-8fad-4352-a4df-7bbbf3635542","added_by":"auto","created_at":"2025-08-08 16:12:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2851226,"visible":true,"origin":"","legend":"\u003cp\u003eSpearman correlation analysis between dietary vitamins and covariates\u003c/p\u003e","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/603d1ab874000d9d8418526e.png"},{"id":88642969,"identity":"1e639c94-4926-4145-8858-aa5f2ae2b21e","added_by":"auto","created_at":"2025-08-08 16:12:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2291756,"visible":true,"origin":"","legend":"\u003cp\u003evariance inflation factor for covariates\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/54ea27cec2626898ba1f2715.png"},{"id":88642975,"identity":"25b36b28-6947-4382-a27d-ffbb63bdafda","added_by":"auto","created_at":"2025-08-08 16:12:47","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":3163448,"visible":true,"origin":"","legend":"\u003cp\u003ePredictor screening results. (A) LASSO regression model screening variable trajectories; (B) Factor screening based on the LASSO regression model, with the left dashed line indicating the best lambda value for the evaluation metrics (lambda. min) and the right dashed line indicating the lambda value for the model where the evaluation metrics are in the range of the best value by one standard error (lambda.1se); (C) \u0026nbsp;\u0026nbsp;Boruta; (D) common predictors between Boruta and LASSO.\u003c/p\u003e","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/0a7883041a77e460afb66559.png"},{"id":88644684,"identity":"ce59b83d-7651-4d45-be8f-e31e12d460bc","added_by":"auto","created_at":"2025-08-08 16:20:46","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2974914,"visible":true,"origin":"","legend":"\u003cp\u003eROC Curves for 10 ML Models\u003c/p\u003e","description":"","filename":"fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/ea3226d4ec4e0f84bf949b9f.png"},{"id":88642964,"identity":"eb1e2158-ec31-4f0b-8552-9d5c9d15248a","added_by":"auto","created_at":"2025-08-08 16:12:47","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1624203,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration curves for 10 predictive models.\u003c/p\u003e","description":"","filename":"fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/a7d7ee2d9de0a3ab5fe01e61.png"},{"id":88644710,"identity":"ac951777-70f8-49ad-a4e4-f081eba5dbdd","added_by":"auto","created_at":"2025-08-08 16:20:47","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":2746196,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curves for 10 forecasting models.\u003c/p\u003e","description":"","filename":"fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/5987ac195e8e3b798066bc70.png"},{"id":88644687,"identity":"e2498d94-6191-4310-b5a1-3063ebf56077","added_by":"auto","created_at":"2025-08-08 16:20:46","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1525821,"visible":true,"origin":"","legend":"\u003cp\u003eInterpretability analysis of logistic regression models. \u003cstrong\u003e(A)\u003c/strong\u003e SHAP dendrogram of features of the logistic regression model. \u003cstrong\u003e(B)\u003c/strong\u003e Importance ranking plot of features of the logistic regression model.\u003c/p\u003e","description":"","filename":"fig8.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/6b00665689898e036900117e.png"},{"id":88645731,"identity":"c2886814-a32f-4e56-8a61-5f42a7fa8a85","added_by":"auto","created_at":"2025-08-08 16:28:46","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":388994,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version.\u003c/p\u003e","description":"","filename":"fig9.png","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/a5a12a0aa41c4e2875de589f.png"},{"id":93649863,"identity":"56f61964-a6a2-4d94-825b-87debc0cd329","added_by":"auto","created_at":"2025-10-16 05:31:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":19328027,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6687344/v1/5fcdf3d4-7023-499b-8e40-495ff05b76bc.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Association Between Dietary Vitamin Intake and Severe Abdominal Aortic Calcification in U.S. Adults: A SHAP-Based Machine Learning Analysis of NHANES 2013–2014","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eSevere abdominal aortic calcification (AAC) serves as a definitive marker of subclinical atherosclerosis and a critical predictor of cardiovascular disease (CVD) (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) .Cardiovascular disease is the leading cause of non-communicable diseases worldwide, imposing a significant burden on families and society(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). In the United States, the incidence of AAC increases with age, and its prevalence continues to rise annually (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Although studies suggest that the exact mechanism of vascular calcification likely involves elements of bone metabolism, our understanding of the mechanisms underlying AAC remains incomplete, and effective prevention and treatment strategies are still lacking(\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Therefore, early detection and a comprehensive understanding of the risk factors associated with AAC are crucial for effective cardiovascular risk assessment and management.\u003c/p\u003e\u003cp\u003eDiet plays a crucial role in vascular health, and previous studies have demonstrated that dietary factors, such as the intake of vitamins, are associated with the incidence of abdominal aortic calcification(\u003cspan additionalcitationids=\"CR8 CR9\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). This association may be attributed to the vitamins' ability to scavenge free radicals and inhibit oxidation. However, existing studies have primarily focused on the individual effects of specific dietary vitamin intakes, and research on the combined effects of dietary and supplemental sources of vitamin intake remains insufficient. Given the potential interactions among different dietary vitamins, examining a single micronutrient in isolation fails to account for the synergistic effects of dietary vitamin intake on the risk of severe abdominal aortic calcification. Therefore, further investigation into the protective role of dietary vitamins in this complex pathology is of significant importance.\u003c/p\u003e\u003cp\u003eOur study aims to utilize the National Health and Nutrition Examination Survey (NHANES) to identify potential associations between dietary vitamin intake and severe abdominal aortic calcification through machine learning (ML) methods. Unlike traditional statistical approaches, ML techniques can manage large and complex datasets while uncovering implicit relationships among various health-related features, thus enabling more accurate predictions of disease risk(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). These early prediction models will aid in controlling the extent of abdominal aortic calcification and enhance public awareness of prevention, thereby making a significant contribution to reducing the socioeconomic burden and adverse public health events associated with this irreversible and severe vascular disease.\u003c/p\u003e"},{"header":"2 Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Study population\u003c/h2\u003e\u003cp\u003eNHANES is a complex, multistage sampling program conducted by the Centers for Disease Control and Prevention (CDC) that serves as the data source for the analysis in this study. The survey received approval from the Institutional Review Board (IRB), and written consent was obtained from all participants. Participants from the 2013\u0026ndash;2014 NHANES were considered candidates for this study. Cases with missing data for either independent or dependent variables were excluded, and the remaining participants were included for further analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Dietary vitamin intake\u003c/h2\u003e\u003cp\u003eThe intake data of 11 dietary vitamins were obtained from NHANES. Participants underwent two 24-hour dietary recall interviews at mobile examination centers, with an interval of 3 to 10 days between the two interviews. The average daily intake of dietary antioxidants was subsequently calculated.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 AAC\u003c/h2\u003e\u003cp\u003eAbdominal aortic calcification (AAC) was defined using the Kauppila scoring system, which employs dual-energy X-ray absorptiometry. The assessment divides the abdominal aortic wall into four segments, with each segment scored from 0 to 6 based on the level of calcium deposition. This results in a cumulative AAC score ranging from 0 to 24. Typically, an AAC score greater than 6 is considered indicative of severe AAC(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Covariates\u003c/h2\u003e\u003cp\u003eDemographic data were obtained using standardized surveys that covered gender, race/ethnicity, educational background, smoking habits, alcohol consumption, poverty income ratio (PIR), body mass index (BMI), and blood pressure (BP). Participants were categorized into the drinking group and the non-drinking group based on their response to the question, \"Do you drink 4/5 cups or more of alcohol per day?\" Smoking status was assessed as never smoker (fewer than 100 cigarettes), former smoker (not currently smoking but having smoked 100 or more cigarettes), or current smoker (having smoked 100 or more cigarettes and currently smoking every day or on some days). The definition of hypertension was based on the question, \"Has your doctor ever told you that you have hypertension?\" The criteria for diagnosing diabetes include any of the following: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) HbA1c\u0026thinsp;\u0026ge;\u0026thinsp;6.5%; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) fasting plasma glucose (FPG)\u0026thinsp;\u0026ge;\u0026thinsp;7.0 mmol/L; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) random plasma glucose (RPG)\u0026thinsp;\u0026ge;\u0026thinsp;11.1 mmol/L or oral glucose tolerance test (OGTT)\u0026thinsp;\u0026ge;\u0026thinsp;11.1 mmol/L; (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) a doctor\u0026rsquo;s diagnosis of diabetes; or (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) currently taking anti-diabetic medications or insulin(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Correlation Analysis and Feature Selection\u003c/h2\u003e\u003cp\u003eMulticollinearity refers to the correlation between two or more variables. In practice, when variables are correlated, interpreting the impact of an independent variable on the dependent variable becomes challenging. Specifically, multiple correlated independent variables jointly influence the dependent variable. Consequently, the presence of multicollinearity reduces the reliability of the estimated coefficients and diminishes the stability and performance of the model. To detect multicollinearity among the included features, we generally use correlation coefficients and variance inflation factors. We consider that multicollinearity exists among independent variables when the correlation coefficient exceeds 0.9 or the variance inflation factor exceeds 10. Since the binary covariates and vitamin concentrations in the statistical model are non-normally distributed, Spearman correlation analysis is employed to detect associations among all independent variables. Additionally, due to the excessive number of independent variables identified by the model, we employed the Boruta algorithm and the Least Absolute Shrinkage and Selection Operator (LASSO) for feature selection.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Machine learning model building\u003c/h2\u003e\u003cp\u003eTo establish a predictive model for severe abdominal aortic calcification and assess its predictive capability, the data were randomly divided into a training set and a test set in a 7:3 ratio. To address the issue of class imbalance in the initial dataset, SMOTE was applied to the training set to enhance the performance of the machine learning (ML) models. The algorithms employed in this study included Neural Networks (NN), Random Forest (RF), Extreme Gradient Boosting (XGB), Gradient Boosting Machine (GBM), Logistic Regression (LR), Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Light Gradient Boosting Machine (LGB), Adaptive Boosting (Adaboost), and Categorical Boosting (CatBoost), totaling ten ML algorithms to predict the risk of severe abdominal aortic calcification. The constructed models were subsequently applied to the test set. Additionally, the project team is developing a new technology for Xiaomi.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Model evaluation and interpretation\u003c/h2\u003e\u003cp\u003eFor the validation set, the performance of the machine learning model was evaluated using several metrics: accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve (AUC). The model exhibiting the best discriminative power was identified. Subsequently, calibration curves and decision curves were plotted to further validate the model's predictive ability and practical application value. The model was interpreted using Shapley Additive Explanations (SHAP).\u003c/p\u003e\u003cp\u003eAll analyses and modeling in this study were conducted using the statistical software R version 4.4.2 and RStudio. All tests were two-sided, and a p-value of less than 0.05 was considered statistically significant.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Baseline characteristics of study populations\u003c/h2\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates the screening process for the study subjects, including a total of 1,070 adults with complete data. Among these, 993 individuals did not have severe abdominal aortic calcification (SAAC), while 137 did. Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e presents the baseline characteristics of the study population. Patients with severe SAAC generally exhibited lower dietary vitamin content compared to those without severe SAAC. Furthermore, comparisons of baseline characteristics between the two groups revealed statistically significant differences in age, ethnicity, BMI, smoking status, education level, hypertension, and diabetes (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eBaseline characteristics of the study population\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCharacteristic\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eNon-Sever AAC\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;933 (89%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSever AAC\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;137 (11%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP -Value\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.00 (11.40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e70.51 (9.42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.388\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eFemale\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e497.00(53.27%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67.00(48.91%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eMale\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e436.00 (46.73%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e70.00 (51.09%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eMexican American\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e121.00 (12.97%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e15.00 (10.95%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eOther Hispanic\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e85 .00(9.11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6 .00(4.38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNon-Hispanic White\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e440.00 (47.16%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e86 .00(62.77%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNon-Hispanic Black\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e186 .00(19.94%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e22.00 (16.06%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eOther\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e101.00 (10.83%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8.00 (5.84%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eBMI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e28.69 (5.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26.91 (4.46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlcohol\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.487\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNo\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e769.00 (84.33%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e109.00 (78.54%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eYes\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e164.00 (15.67%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e28.00 (21.46%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eSmoke\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNever\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e466.00 (50.54%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e45.00 (36.72%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eformer\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e283.00 (31.28%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e61.00 (45.22%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eCurrent\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e184.00 (18.18%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e31.00 (18.06%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eEducational\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.021\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026gt;=High school\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e751.00 (86.99%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e98.00 (74.63%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003e\u0026lt;High school\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e182.00 (13.01%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e39.00 (25.37%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePIR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.79 (1.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.65 (1.59)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.324\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypertension\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNo\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e518.00 (58.32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e43.00 (33.43%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eYes\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e415.00 (41.68%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e94.00 (66.57%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiabetes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eNo\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e745.00 (84.31%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e89.00 (69.74%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eYes\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e188.00 (15.69%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48.00 (30.26%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003efolicacid\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e233.89 (137.58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e183.50 (80.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitamin A\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e629.44 (454.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e535.26 (328.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.02\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitaminB1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.59 (0.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.37 (0.58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitaminB2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.10 (1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.85 (0.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.006\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitaminB6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.11 (1.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.75 (0.83)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitaminB12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.84 (3.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.94 (2.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitamin C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e80.59 (73.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e77.28 (81.36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.628\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitamin D\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5.08 (5.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.16 (3.31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.046\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitamin E\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.79 (3.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.37 (1.49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.122\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eVitamin K\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e130.73 (149.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e91.43 (75.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.003\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNiacin\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e25.50 (12.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e21.79 (9.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Correlation Analysis and Feature Selection\u003c/h2\u003e\n \u003cp\u003eWe explored the correlations between dietary vitamins and other independent variables using Spearman correlation analysis. The results of these correlations, presented in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, show the correlation coefficients (r) between dietary vitamins and participants\u0026apos; baseline characteristics. Most vitamins exhibited varying degrees of correlation, with the most significant correlation observed between niacin and vitamin B6 (r\u0026thinsp;=\u0026thinsp;0.81). This was followed by strong correlations between niacin and vitamin B1 (r\u0026thinsp;=\u0026thinsp;0.71) and vitamin B2 (r\u0026thinsp;=\u0026thinsp;0.68). In this study, we defined a correlation coefficient greater than 0.9 as indicative of multicollinearity. Furthermore, the variance inflation factor (VIF) values of the included covariates were all below 10 (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e), indicating the absence of multicollinearity among the variables.\u003c/p\u003e\n \u003cp\u003eWe divided the data into a training group consisting of 749 cases and a testing group comprising 321 cases, maintaining a 7:3 ratio. Statistical analysis revealed no significant differences between the two groups (Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). LASSO regression, a shrinkage estimation method, achieves variable selection and complexity adjustment by formulating an optimization objective function that includes a penalty term. In this study, LASSO regression was employed to identify characteristic factors such as folic acid, diabetes, smoking, vitamin A, age, and BMI (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA, B). The Boruta algorithm, an extension of the RF algorithm, accurately estimates the importance of each feature, thereby identifying the true set of features. We utilized the Boruta algorithm to identify 14 key factors, including age, BMI, diabetes, hypertension, folic acid, vitamin A, vitamin B, vitamin C, vitamin D, and niacin (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eC). Through comparative analysis of the screening results from LASSO regression and the Boruta algorithm, we determined the subset of common feature variables selected by both methods, ultimately narrowing down to five feature variables: diabetes, folic acid, vitamin A, age, and BMI (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eD).\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\u003cbr\u003e\u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003etest\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003etrain\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep.overall\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eN\u0026thinsp;=\u0026thinsp;321\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003eN\u0026thinsp;=\u0026thinsp;749\u003c/em\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSEQN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e78697 (2883)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e78704 (2958)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.969\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eResult:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.281\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e274 (85.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e659 (88.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47 (14.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e90 (12.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSex:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.400\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e176 (54.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e388 (51.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e145 (45.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e361 (48.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRace:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.439\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e43 (13.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e93 (12.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 (6.23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e71 (9.48%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e165 (51.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e361 (48.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e59 (18.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e149 (19.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34 (10.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e75 (10.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEducational:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.717\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e69 (21.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e152 (20.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e252 (78.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e597 (79.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAlcohol:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.289\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e270 (84.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e608 (81.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e51 (15.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e141 (18.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSmoke:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.351\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e147 (45.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e364 (48.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e120 (37.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e224 (29.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54 (16.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e161 (21.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHypertension:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.979\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e169 (52.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e392 (52.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e152 (47.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e357 (47.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDiabetes:\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.394\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e256 (79.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e578 (77.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e65 (20.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e171 (22.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e58.8 (12.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e58.7 (12.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.887\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePIR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.82 (1.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.76 (1.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.595\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e28.1 (5.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e28.6 (5.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.165\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003efolicacid\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e227 (124)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e228 (136)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.902\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitaminA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e614 (456)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e619 (436)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.880\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitaminB1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.57 (0.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.56 (0.70)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.985\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitaminB2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.10 (1.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.06 (0.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.630\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitaminB6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.17 (1.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.02 (1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.058\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitaminB12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4.96 (4.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4.62 (3.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.200\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitamin C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e89.4 (79.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e76.2 (72.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.051\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitamin D\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5.30 (5.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4.81 (4.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.153\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitamin E\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.75 (2.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.73 (3.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.933\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVitamin K\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e119 (142)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e129 (143)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.309\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNiacin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25.9 (14.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.7 (11.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.178\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003ePredictor screening results. (A) LASSO regression model screening variable trajectories; (B) Factor screening based on the LASSO regression model, with the left dashed line indicating the best lambda value for the evaluation metrics (lambda. min) and the right dashed line indicating the lambda value for the model where the evaluation metrics are in the range of the best value by one standard error (lambda.1se); (C) Boruta; (D) common predictors between Boruta and LASSO.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Evaluation of ML Models\u003c/h2\u003e\n \u003cp\u003eBased on the training data, we plotted the ROC curve (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e), which revealed that the Logistic model exhibited the best predictive performance, achieving an AUC of 0.851, indicative of high prediction accuracy. By integrating metrics such as accuracy, precision, recall, and F1 score, we conducted a comprehensive evaluation of the model\u0026apos;s performance, concluding that the Logistic model outperformed the others (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Consequently, the Logistic model was selected for further analysis.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eML performance index\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eModel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAUC\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePrecision\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eF1\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLogistic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.851(0.792\u0026ndash;0.909)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.787\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.789\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.352\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.485\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.833(0.772\u0026ndash;0.895)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.234\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.195\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.036\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.061\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGBM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.807(0.738\u0026ndash;0.876)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.716\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.805\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.703\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.284\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eKNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.806(0.744\u0026ndash;0.868)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.566\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.976\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.505\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.225\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.365\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.792(0.723\u0026ndash;0.862)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.713\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.805\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.699\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.282\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.418\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eXGB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.793(0.713\u0026ndash;0.873)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.803\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.732\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.814\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.366\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.488\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eKNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.806(0.744\u0026ndash;0.868)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.566\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.976\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.505\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.225\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.365\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAdaboost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.743(0.678\u0026ndash;0.807)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.678\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.829\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.656\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.262\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.398\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLGB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.666(0.554\u0026ndash;0.777)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.719\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.659\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.728\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.262\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.375\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCatBoost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.839(0.775\u0026ndash;0.902)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.691\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.951\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.652\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.287\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.441\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eSince the discriminative performance had minimal impact on the comparative predictive performance of the models, this study proceeded to evaluate the calibration degree of each model through calibration curves. The calibration curve of the Logistic model in the validation cohort exhibited good consistency, closely following the ideal diagonal line (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e), which indicates high predictive accuracy. Furthermore, the Decision Curve Analysis (DCA) in the validation cohort demonstrated that across a wide range of threshold probabilities, the net benefit of this model consistently exceeded that of the two extreme strategies (treat-all and treat-none), suggesting its potential clinical utility (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e). In summary, our evaluation results indicate that our machine learning model can effectively identify and benefit patients with severe abdominal aortic calcification.\u003c/p\u003e\n \u003cp\u003eFinally, we plotted the SHAP diagram (Fig. \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e) to illustrate the impact of various features on the model\u0026apos;s output. Age had the most significant influence, with higher values contributing positively and lower values contributing negatively; notably, a considerable proportion of individuals had a negative impact on the prediction. Following this, diabetes also exerted a negative influence, predominantly among individuals with lower values. Vitamin A ranked fourth in its effect on the model\u0026apos;s output, where higher values primarily contributed negatively and lower values contributed positively. Folic acid had the least impact on the model\u0026apos;s predictions, with lower values mainly contributing positively. The results indicated that the two most significant potential key heavy metals associated with SAAC risk are Vitamin A and folic acid.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eConstruction of nomograms\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eThis study integrated five significant predictive variables\u0026mdash;Diabetes, folic acid, Vitamin A, Age, and BMI\u0026mdash;to intuitively assess the risk of severe abdominal aortic calcification. Figure \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003e presents the nomogram, which offers a rapid and easily interpretable risk assessment.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eSevere abdominal aortic calcification poses a significant threat to human health. In previous studies, we identified a correlation between various dietary vitamins and abdominal aortic calcification, with the majority acting as protective factors against its occurrence. To date, we have not discovered any predictive models for severe abdominal aortic calcification. Dietary vitamin intake, as a modifiable lifestyle factor, exhibits not only long-term stability but also clear value for intervention guidance. Particularly at the public health level, prediction models based on nutritional behaviors possess stronger generalizability and preventive potential. Therefore, this study aims to address this gap by developing a predictive model for severe abdominal aortic calcification based on dietary vitamins, providing a new perspective for early risk assessment and potential dietary interventions. This study explores the complex relationship between dietary vitamins and severe abdominal aortic calcification (SAAC). Among the ten machine learning algorithms, the Logistic model excelled in identifying potential risks of SAAC (AUC\u0026thinsp;=\u0026thinsp;0.851). SHAP values indicated that, among dietary vitamins, vitamin A contributed the most to the machine learning model. Age was the most significant feature among all characteristics, while folic acid had the least impact on the machine learning model. This model serves as a valuable auxiliary tool for clinical decision-making.\u003c/p\u003e\u003cp\u003eThis study indicates that among dietary vitamins, Vitamin A exhibits the strongest correlation with SAAC. In a study focused on identifying dietary components associated with abdominal aortic calcification, data on 35 macronutrients and micronutrients were analyzed, revealing that Vitamin A concentration is inversely correlated with abdominal aortic calcification(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Currently, the literature on the association between vitamin A intake and aortic arch calcification (AAC) remains limited. Due to its antioxidant properties, vitamin A supplementation may be linked to protective effects at the arterial level(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). However, considering the potential toxicity of high doses of vitamin A, further research is essential to investigate its application as a protective measure against AAC.\u003c/p\u003e\u003cp\u003eIn this study, we found that age is a significant predictor of SAAC, which aligns with common understanding. The atherosclerotic process begins in childhood and may progress to the formation of advanced atherosclerotic plaques, with some lesions undergoing calcification. Multiple studies have explored the relationship between aortic calcification and age, all concluding that abdominal aortic calcification is positively correlated with age(\u003cspan additionalcitationids=\"CR19 CR20 CR21\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Current literature searches have not identified any studies demonstrating a negative or zero correlation with age, suggesting that age is a significant determinant of the presence and severity of abdominal aortic calcification. However, diabetes mellitus is regarded as a critical factor in this study, as it is associated with arterial calcification, including both intimal arterial calcification and atherosclerotic intimal calcification. Multiple studies support the relationship between aortic calcification and diabetes (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). In contrast, Matsushita et al.(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) did not find a correlation between diabetes and the incidence or severity of calcification when evaluating a male cohort; however, this may be attributed to the small sample size of the study. Moreover, the association between BMI and SAAC should not be overlooked. Recent studies suggest that lean body mass may be a risk factor for AAC(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e), and their findings align with those of the current study. Additionally, visceral fat accumulation has been shown to be closely associated with dyslipidemia, insulin resistance, diabetes, and hypertension, all of which increase the risk of cardiovascular disease(\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). This discrepancy may be due to BMI's limitations in accurately reflecting fat distribution(\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFolic acid contributes the least to the machine learning (ML) model. Previous studies have demonstrated that folic acid serves as a protective factor against severe aortic arch calcification (AAC) formation, with the benefits being most pronounced in individuals who have adequate folic acid intake, particularly those with a daily intake exceeding 280.126 mg (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eA potential contributing factor to this association is the regulation of homocysteine levels. Research indicates that folic acid supplementation effectively reduces hyperhomocysteinemia, which is a recognized trigger for vascular dysfunction and an increased risk of cardiovascular disease (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Additionally, oxidative stress can lead to vascular calcification, and folic acid deficiency exacerbates oxidative stress and various aspects of metabolic syndrome, both of which are associated with an elevated risk of diabetes and cardiovascular diseases(\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). Although there has been considerable speculation regarding the role of folic acid in arterial calcification, the specific mechanisms through which its dietary intake influences the occurrence of severe AAC still require in-depth exploration.\u003c/p\u003e\u003cp\u003eHowever, this study has several limitations. Firstly, while the LR model demonstrated strong predictive performance, its generalizability remains uncertain due to the lack of external validation. Secondly, the manually selected features may introduce slight bias, suggesting that advanced algorithms could save time and costs for researchers by utilizing automatic feature selection techniques. Thirdly, dietary vitamin intake is often associated with geographical location; however, this variable was not included in the model. These factors should be considered in future research. Finally, as a cross-sectional study, this research cannot verify causal associations due to the absence of time series data, and the results should be interpreted with caution.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eThis study constructed a machine learning (ML) model to explore the association between the risk of SAAC and dietary vitamins, while also incorporating demographic data. The logistic regression (LR) algorithm demonstrated superior performance compared to the other nine algorithms assessed. Among the eleven vitamins examined, vitamin A emerged as the most significant factor in the ML model, whereas folate exhibited a weaker association with SAAC than vitamin A. Among all independent variables, age was identified as the most crucial factor in this model. Furthermore, diabetic patients exhibited a higher risk of developing SAAC. In contrast to obese individuals, lean body mass may pose a risk factor for AAC. Future research should focus on developing more advanced algorithms to validate these findings and adjust relevant parameters to enhance the model's accuracy.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSevere abdominal aortic calcification\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSAAC\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eMchine learning\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eML\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNational Health and Nutrition Examination Survey\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNHANES\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLeast Absolute Shrinkage and Selection Operator\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLASSO\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSynthetic Minority Over-sampling Technique\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSMOTE\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNeural Networks\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNN\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eRandom Forests\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eExtreme Gradient Boosting\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eXGB\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGradient Boosting Machines\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eGBM\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLogistic Regression\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLR\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSupport Vector Machine\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSVM\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ek-Nearest Neighbors\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eKNN\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLight Gradient Boosting Machine\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLGB\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eAdaptive Boosting\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eAdaboost\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ethe area under the Receiver Operating Characteristic curve\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eThe Shapley Additive Explanations\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSHAP\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCrdiovascular disease\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eCVD\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTe Centers for Disease Control and Prevention\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eCDC\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eThe Institutional Review Board\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eIRB\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePoverty income ratio\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ePIR\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eBody mass index\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eBMI\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eBlood pressure\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eBP\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eThe variance inflation factor\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eVIF\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDecision Curve Analysis\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eDCA\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research involving human subjects received approval from the Institutional Review Board of the National Center for Health Statistics. The study was conducted in accordance with local legislation and institutional requirements, and adhered to the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants for their involvement in this research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors of this study have fully discussed and unanimously agreed to submit this manuscript for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe complete list of datasets supporting the findings of this study is available at the following URL:https://www.cdc.gov/nchs/nhanes/。\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author declares that this research was conducted without any commercial or financial relationships that could be perceived as a potential conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author(s) declare that no financial support was received for the research and/or publication of this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDW: Data Management, Writing-Original Draft.\u0026nbsp;ZL: Software, validation, writing-review and edit.\u0026nbsp;SG: Funding, supervision, writing-reviewing and editing.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAllam AHA, Thompson RC, Eskander MA, Mandour Ali MA, Sadek A, Rowan CJ, Sutherland ML, Sutherland JD, Frohlich B, Michalik DE, Finch CE, Narula J, Thomas GS, Samuel Wann L. Is coronary calcium scoring too late? Total body arterial calcium burden in patients without known CAD and normal MPI. J nuclear cardiology: official publication Am Soc Nuclear Cardiol. 2018;25:1990\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12350-017-0925-9\u003c/span\u003e\u003cspan address=\"10.1007/s12350-017-0925-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, Barengo NC, Beaton AZ, Benjamin EJ, Benziger CP, Bonny A, Brauer M, Brodmann M, Cahill TJ, Carapetis J, Catapano AL, Chugh SS, Cooper LT, Coresh J, Criqui M, DeCleene N, Eagle KA, Emmons-Bell S, Feigin VL, Fern\u0026aacute;ndez-Sol\u0026agrave; J, Fowkes G, Gakidou E, Grundy SM, He FJ, Howard G, Hu F, Inker L, Karthikeyan G, Kassebaum N, Koroshetz W, Lavie C, Lloyd-Jones D, Lu HS, Mirijello A, Temesgen AM, Mokdad A, Moran AE, Muntner P, Narula J, Neal B, Ntsekhe M, de Moraes G, Otto C, Owolabi M, Pratt M, Rajagopalan S, Reitsma M, Ribeiro ALP, Rigotti N, Rodgers A, Sable C, Shakil S, Sliwa-Hahnle K, Stark B, Sundstr\u0026ouml;m J, Timpel P, Tleyjeh IM, Valgimigli M, Vos T, Whelton PK, Yacoub M, Zuhlke L, Murray C, Fuster V. Global Burden of Cardiovascular Diseases and Risk Factors, 1990\u0026ndash;2019: Update From the GBD 2019 Study. J Am Coll Cardiol. 2020;76:2982\u0026ndash;3021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jacc.2020.11.010\u003c/span\u003e\u003cspan address=\"10.1016/j.jacc.2020.11.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWong ND, Lopez VA, Allison M, Detrano RC, Blumenthal RS, Folsom AR, Ouyang P, Criqui MH. Abdominal aortic calcium and multi-site atherosclerosis: the Multiethnic Study of Atherosclerosis. \u003cem\u003eAtherosclerosis\u003c/em\u003e.(2011)214:436\u0026thinsp;\u0026ndash;\u0026thinsp;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.atherosclerosis.2010.09.011\u003c/span\u003e\u003cspan address=\"10.1016/j.atherosclerosis.2010.09.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePan W, Jie W, Huang H. Vascular calcification: Molecular mechanisms and therapeutic interventions. MedComm (2023)4:e200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/mco2.200\u003c/span\u003e\u003cspan address=\"10.1002/mco2.200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDoherty TM, Fitzpatrick LA, Inoue D, Qiao JH, Fishbein MC, Detrano RC, Shah PK, Rajavashisth TB. Molecular, endocrine, and genetic mechanisms of arterial calcification. Endocr Rev. 2004;25:629\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1210/er.2003-0015\u003c/span\u003e\u003cspan address=\"10.1210/er.2003-0015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcCullough PA, Soman S. Cardiovascular calcification in patients with chronic renal failure: are we on target with this risk factor? Kidney Int Supplement. 2004;S18\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1523-1755.2004.09008.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1523-1755.2004.09008.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi H, Li R, Gong C, Wu Z, Jia Q. The relationship between dietary vitamin B1 intake and severe abdominal aortic calcification among the general population in the United States. Front Cardiovasc Med. 2024;11:1415151. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fcvm.2024.1415151\u003c/span\u003e\u003cspan address=\"10.3389/fcvm.2024.1415151\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEl Maghraoui A, Hamza T, Sadni S, El Maataoui A, Majjad A, Rezqi A, Ouzzif Z, Mounach A. Vitamin D status and abdominal aortic calcification in postmenopausal women. J Bone Miner Metab. 2018;36:229\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00774-017-0832-9\u003c/span\u003e\u003cspan address=\"10.1007/s00774-017-0832-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJia J, Zhang J, He Q, Wang M, Liu Q, Wang T, Chen X, Wang W, Xu H. Association between dietary vitamin C and abdominal aortic calcification among the US adults. Nutr J. 2023;22:58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12937-023-00889-y\u003c/span\u003e\u003cspan address=\"10.1186/s12937-023-00889-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOyama S, Okamoto N, Koide S, Hayashi H, Nakai S, Takahashi K, Inaguma D, Hasegawa M, Toyama H, Sugiyama S, Yuzawa Y, Tsuboi N. Vitamin K2 supplementation and the progression of abdominal aortic calcification in dialysis patients. Fujita Med J. 2021;7:136\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.20407/fmj.2020-020\u003c/span\u003e\u003cspan address=\"10.20407/fmj.2020-020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRajula HSR, Verlato G, Manchia M, Antonucci N, Fanos V. Comparison of Conventional Statistical Methods with Machine Learning in Medicine: Diagnosis, Drug Development, and Treatment. Med (Kaunas Lithuania). 2020;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/medicina56090455\u003c/span\u003e\u003cspan address=\"10.3390/medicina56090455\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi X, Zhao Y, Zhang D, Kuang L, Huang H, Chen W, Fu X, Wu Y, Li T, Zhang J, Yuan L, Hu H, Liu Y, Zhang M, Hu F, Sun X, Hu D. Development of an interpretable machine learning model associated with heavy metals' exposure to identify coronary heart disease among US adults via SHAP: Findings of the US NHANES from 2003 to 2018. Chemosphere. 2023;311:137039. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.chemosphere.2022.137039\u003c/span\u003e\u003cspan address=\"10.1016/j.chemosphere.2022.137039\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXie R, Liu X, Wu H, Liu M, Zhang Y. Associations between systemic immune-inflammation index and abdominal aortic calcification: Results of a nationwide survey. Nutr metabolism Cardiovasc diseases: NMCD. 2023;33:1437\u0026ndash;43. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.numecd.2023.04.015\u003c/span\u003e\u003cspan address=\"10.1016/j.numecd.2023.04.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXiao Y, Quan Y. Associations between dietary fatty acids intake and abdominal aortic calcification: a national population-based study. Lipids Health Dis. 2024;23:73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12944-024-02059-3\u003c/span\u003e\u003cspan address=\"10.1186/s12944-024-02059-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYin J, Zheng C, Lin X, Huang C, Hu Z, Lin S, Qu Y. The potential of the serum uric acid to high-density lipoprotein cholesterol ratio as a predictive biomarker of diabetes risk: a study based on NHANES 2005\u0026ndash;2018. Front Endocrinol. 2024;15:1499417. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fendo.2024.1499417\u003c/span\u003e\u003cspan address=\"10.3389/fendo.2024.1499417\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi W, Huang G, Tang N, Lu P, Jiang L, Lv J, Qin Y, Lin Y, Xu F, Lei D. Identification of dietary components in association with abdominal aortic calcification. Food Funct. 2023;14:8383\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1039/d3fo02920d\u003c/span\u003e\u003cspan address=\"10.1039/d3fo02920d\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eda Rocha RF, de Oliveira MR, Schonhofen P, De Bastiani MA, Schnorr CE, Klamt F, Dal Pizzol F, Moreira JC. Vitamin A supplementation for different periods alters rat vascular redox parameters. J Physiol Biochem. 2010;66:351\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s13105-010-0041-7\u003c/span\u003e\u003cspan address=\"10.1007/s13105-010-0041-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllison MA, Criqui MH, Wright CM. Patterns and risk factors for systemic calcified atherosclerosis. \u003cem\u003eArteriosclerosis, thrombosis, and vascular biology\u003c/em\u003e.(2004)24:331-6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/01.ATV.0000110786.02097.0c\u003c/span\u003e\u003cspan address=\"10.1161/01.ATV.0000110786.02097.0c\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKiel DP, Kauppila LI, Cupples LA, Hannan MT, O'Donnell CJ, Wilson PW. Bone loss and the progression of abdominal aortic calcification over a 25 year period: the Framingham Heart Study. Calcif Tissue Int. 2001;68:271\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/bf02390833\u003c/span\u003e\u003cspan address=\"10.1007/bf02390833\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiwa Y, Tsushima M, Arima H, Kawano Y, Sasaguri T. Pulse pressure is an independent predictor for the progression of aortic wall calcification in patients with controlled hyperlipidemia. \u003cem\u003eHypertension (Dallas, Tex\u003c/em\u003e: 1979).(2004)43:536\u0026thinsp;\u0026ndash;\u0026thinsp;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/01.HYP.0000117153.48029.d1\u003c/span\u003e\u003cspan address=\"10.1161/01.HYP.0000117153.48029.d1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eO'Donnell CJ, Chazaro I, Wilson PW, Fox C, Hannan MT, Kiel DP, Cupples LA. Evidence for heritability of abdominal aortic calcific deposits in the Framingham Heart Study. \u003cem\u003eCirculation\u003c/em\u003e.(2002)106:337\u0026thinsp;\u0026ndash;\u0026thinsp;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/01.cir.0000022663.26468.5b\u003c/span\u003e\u003cspan address=\"10.1161/01.cir.0000022663.26468.5b\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eReaven PD, Sacks J. Reduced coronary artery and abdominal aortic calcification in Hispanics with type 2 diabetes. Diabetes Care. 2004;27:1115\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2337/diacare.27.5.1115\u003c/span\u003e\u003cspan address=\"10.2337/diacare.27.5.1115\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatsushita M, Nishikimi N, Sakurai T, Nimura Y. Relationship between aortic calcification and atherosclerotic disease in patients with abdominal aortic aneurysm. Int Angiol. 2000;19:276\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYuan M, Hsu FC, Bowden DW, Xu J, Carrie Smith S, Wagenknecht LE, Comeau ME, Divers J, Register TC, Jeffrey Carr J, Langefeld CD, Freedman BI. Relationships between measures of adiposity with subclinical atherosclerosis in patients with type 2 diabetes. \u003cem\u003eObesity (Silver Spring, Md)\u003c/em\u003e.(2016)24:1810-8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/oby.21540\u003c/span\u003e\u003cspan address=\"10.1002/oby.21540\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRodr\u0026iacute;guez AJ, Scott D, Khan B, Khan N, Hodge A, English DR, Giles GG, Ebeling PR. Low Relative Lean Mass is Associated with Increased Likelihood of Abdominal Aortic Calcification in Community-Dwelling Older Australians. Calcif Tissue Int. 2016;99:340\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00223-016-0157-z\u003c/span\u003e\u003cspan address=\"10.1007/s00223-016-0157-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBray GA, Heisel WE, Afshin A, Jensen MD, Dietz WH, Long M, Kushner RF, Daniels SR, Wadden TA, Tsai AG, Hu FB, Jakicic JM, Ryan DH, Wolfe BM, Inge TH. The Science of Obesity Management: An Endocrine Society Scientific Statement. Endocr Rev. 2018;39:79\u0026ndash;132. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1210/er.2017-00253\u003c/span\u003e\u003cspan address=\"10.1210/er.2017-00253\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, Vasan RS, Murabito JM, Meigs JB, Cupples LA, D'Agostino RB, Sr., O'Donnell CJ. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116:39\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/circulationaha.106.675355\u003c/span\u003e\u003cspan address=\"10.1161/circulationaha.106.675355\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBritton KA, Massaro JM, Murabito JM, Kreger BE, Hoffmann U, Fox CS. Body fat distribution, incident cardiovascular disease, cancer, and all-cause mortality. J Am Coll Cardiol. 2013;62:921\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jacc.2013.06.027\u003c/span\u003e\u003cspan address=\"10.1016/j.jacc.2013.06.027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang K, Chen J, Chen B, Han Y, Cai T, Zhao J, Gu Z, Gao M, Hou Z, Yu X, Gu F, Gao Y, Hu R, Xie J, Liu T, Cui D, Li B. Association between dietary folate intake and severe abdominal aorta calcification in adults: A cross-sectional analysis of the national health and nutrition examination survey. Diabetes Vasc Dis Res. 2024;21:14791641241246555. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/14791641241246555\u003c/span\u003e\u003cspan address=\"10.1177/14791641241246555\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi T, Dong G, Kang Y, Zhang M, Sheng X, Wang Z, Liu Y, Kong N, Sun H. Increased homocysteine regulated by androgen activates autophagy by suppressing the mammalian target of rapamycin pathway in the granulosa cells of polycystic ovary syndrome mice. Bioengineered. 2022;13:10875\u0026ndash;88. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/21655979.2022.2066608\u003c/span\u003e\u003cspan address=\"10.1080/21655979.2022.2066608\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSung ML, Wu CC, Chang HI, Yen CK, Chen HJ, Cheng JC, Chien S, Chen CN. Shear stress inhibits homocysteine-induced stromal cell-derived factor-1 expression in endothelial cells. Circul Res. 2009;105:755\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/circresaha.109.206524\u003c/span\u003e\u003cspan address=\"10.1161/circresaha.109.206524\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYamada S, Taniguchi M, Tokumoto M, Toyonaga J, Fujisaki K, Suehiro T, Noguchi H, Iida M, Tsuruya K, Kitazono T. The antioxidant tempol ameliorates arterial medial calcification in uremic rats: important role of oxidative stress in the pathogenesis of vascular calcification in chronic kidney disease. J bone mineral research: official J Am Soc Bone Mineral Res. 2012;27:474\u0026ndash;85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/jbmr.539\u003c/span\u003e\u003cspan address=\"10.1002/jbmr.539\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePravenec M, Kozich V, Krijt J, Sokolov\u0026aacute; J, Z\u0026iacute;dek V, Landa V, Sim\u0026aacute;kov\u0026aacute; M, Mlejnek P, Silhavy J, Oliyarnyk O, Kazdov\u0026aacute; L, Kurtz TW. Folate deficiency is associated with oxidative stress, increased blood pressure, and insulin resistance in spontaneously hypertensive rats. Am J Hypertens. 2013;26:135\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/ajh/hps015\u003c/span\u003e\u003cspan address=\"10.1093/ajh/hps015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"SAAC, Dietary vitamin, Machine learning, NHANES, AUC","lastPublishedDoi":"10.21203/rs.3.rs-6687344/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6687344/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eSevere abdominal aortic calcification (SAAC) significantly impacts families and society. Although vitamin intake is closely linked to SAAC, large-scale model-based dietary studies are scarce. This study compares machine learning models to analyze this association and support dietary strategies for SAAC prevention.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eThere are 10 ways to build ML models. The best model for further analysis was selected based on accuracy, area under the subject operating characteristic curve (AUC), precision, recall, and F1 score. Shapley Additive Explanations (SHAP) method was used to explain the contribution of variables to ML models.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eLogistic Regression (LR) exhibited the best performance in exploring the association between dietary vitamins and SAAC, with an AUC of 0.851. The SHAP values indicate that among dietary vitamins, vitamin A has the greatest contribution to the machine learning (ML) model. Age is the most important feature among all characteristics, while folate has the least impact on the ML model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e LR algorithm performed best, vitamin A was the most significant factor, folic acid correlation was weak.The model is helpful for early screening and intervention of SAAC and improves patient prognosis.\u003c/p\u003e","manuscriptTitle":"Association Between Dietary Vitamin Intake and Severe Abdominal Aortic Calcification in U.S. Adults: A SHAP-Based Machine Learning Analysis of NHANES 2013–2014","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-08 16:12:40","doi":"10.21203/rs.3.rs-6687344/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"30294ddc-2070-41f0-9926-b3b0b14e8974","owner":[],"postedDate":"August 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-10-16T05:23:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-08 16:12:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6687344","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6687344","identity":"rs-6687344","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}