Exploration and Analysis of Risk Factors for Coronary Artery Disease with Type 2 Diabetes Based on SHAP Explainable Machine Learning Algorithm

Exploration and Analysis of Risk Factors for Coronary Artery Disease with Type 2 Diabetes Based on SHAP Explainable Machine Learning Algorithm | 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 Article Exploration and Analysis of Risk Factors for Coronary Artery Disease with Type 2 Diabetes Based on SHAP Explainable Machine Learning Algorithm Dandan Tang, Fengwei Liang, Xingli Gu, Yuanyuan Jin, Xuanjie Hu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6492298/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 12 Aug, 2025 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Background: Type 2 diabetes (T2DM) significantly elevates coronary heart disease (CHD) risk. This study leverages interpretable machine learning (ML) to identify risk factors for CHD with T2DM, enhancing clinical decision-making. Methods: Clinical data from 5,681 cardiovascular patients (4,396 CHD; 1,285 CHD+T2DM) hospitalized between 2001-2018 were analyzed. The SMOTENC algorithm addressed dataset imbalance. Predictive variables were selected via univariate analysis and Lasso regression. Five ML models (logistic regression, Lasso regression, KNN, SVM, XGBoost) were developed and validated using accuracy, sensitivity, specificity, ROC, and decision curve analysis. SHAP values interpreted model outputs. Results: Data were split into training (n=3,977) and validation (n=1,704) sets. Lasso regression identified 25 predictive variables. XGBoost achieved superior performance (highest accuracy: 0.89; AUC: 0.93) and net benefit in decision curves. SHAP analysis revealed diabetes duration, blood glucose (BG), prothrombin time (PT), and glycated hemoglobin (HbA1c) as primary risk factors. Positive urine glucose and elevated low-density lipoprotein also contributed significantly. Conclusion: Diabetes history, BG, HbA1c, and PT are critical risk factors for CHD-T2DM comorbidity. Prioritizing monitoring of these parameters and implementing targeted interventions may mitigate risk. The XGBoost-SHAP framework provides an interpretable tool for clinical risk stratification. Biological sciences/Biological techniques/Biological models Health sciences/Cardiology Health sciences/Diseases Coronary heart disease combined with type 2 diabetes Machine learning SHAP Imbalance processing Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction With the accelerating aging of China's population, the phenomenon of multimorbidity in the elderly is becoming increasingly prominent. Multimorbidity refers to the simultaneous presence of two or more chronic diseases or conditions [ 1 ] . Among individuals aged 65 and above, the prevalence of multimorbidity reaches 64.7% [ 2 ] .In elderly individuals, the coexistence of coronary heart disease(Coronary Heart Disease,CHD)and diabetes (Diabetes Mellitus,DM) is one of the most common conditions. Coronary heart disease is a key condition within the spectrum of heart diseases, fundamentally caused by organic obstruction or narrowing of the coronary arteries, leading to myocardial ischemia, hypoxia, and even necrosis. Therefore, it is often referred to as ischemic heart disease [ 3 ] . Clinical manifestations include angina, arrhythmia, myocardial infarction, and even sudden death [ 4 ] . A major and modifiable risk factor for coronary heart disease that can be prevented at the population level is hyperglycemia or diabetes [ 5 ] . In recent years, type 2 diabetes has become one of the most important complications of coronary heart disease, with its incidence gradually increasing [ 6 ] , and it is also associated with patient mortality [ 7 ] . In terms of diagnosis, conventional diagnostic techniques include coronary angiography, coronary CT angiography (CTA), electrocardiography (ECG), and cardiac ultrasound. However, these methods require large specialized equipment and trained professionals, making them costly and inconvenient. Therefore, developing low-cost, convenient, and effective non-invasive detection methods is crucial for the early diagnosis of coronary heart disease combined with type 2 diabetes and is expected to significantly reduce patient mortality. Data indicate that glucose metabolism disorders are common among patients undergoing coronary angiography. Among 1,040 patients with coronary heart disease, 62.2% exhibited glucose metabolism abnormalities. The integrated management of coronary heart disease and type 2 diabetes, along with identifying patients at risk for various comorbidities, is a high priority in clinical practice [ 8 ] . Currently, machine learning algorithms have been shown to be very useful in predicting cardiovascular diseases [ 9 ] . In the medical field, the application of machine learning is infiltrating every aspect of clinical practice at an unprecedented pace, from clinical data preprocessing to patient stratification and personalized treatment plans, with its influence becoming increasingly prominent. Specifically, machine learning plays a crucial role in disease diagnosis, treatment risk assessment, drug production, and medical data analysis [ 10 ] . Currently, there is no specific model for predicting the risk of diabetes in coronary heart disease patients.This study uses machine learning algorithms to develop a clinical risk prediction model for coronary heart disease combined with type 2 diabetes.By deeply mining and integrating clinical data from patients with coronary heart disease and type 2 diabetes, and systematically analyzing the key factors contributing to the disease, this study provides solid clinical evidence for early intervention and treatment. The introduction of machine learning models will enable more accurate individual risk assessments and bring new perspectives and possibilities to the development of management strategies for this disease, showcasing broad clinical application potential. Materials and Methods Data and Patients This study selected detailed clinical data from a total of 29,960 cardiovascular disease patients admitted to the First Affiliated Hospital of Xinjiang Medical University between 2001 and 2018 as the research subjects. Based on the following inclusion and exclusion criteria, CHD patients and CHD-DM2 patients were selected. The inclusion criteria are as follows: CHD patients: Diagnosed with CHD through coronary angiography (CAG) or coronary CT angiography (CTA). Have clear clinical symptoms such as angina or other manifestations of ischemic heart disease. Aged 18 years or older. CHD-DM2 patients: Meet all the inclusion criteria for CHD mentioned above. Diagnosed with type 2 diabetes (based on C-peptide levels, insulin autoantibody testing, or age of onset of diabetes, among other indicators). Complete records of blood glucose control, with data on glycated hemoglobin (HbA1c) monitoring. Exclusion criteria: Missing data: Patients with missing key clinical data (such as diagnostic records) or those with obvious errors. Non-CHD patients: Patients who have not been diagnosed as CHD. Only other types of cardiovascular diseases (such as hypertension, heart failure, arrhythmia, myocardial infarction, etc.) are present, but no CHD diagnosis. Patients with other serious systemic diseases: Patients with severe liver or kidney dysfunction or other systemic diseases that may affect the study results. Patients with malignant tumors who are undergoing chemotherapy or radiotherapy. Through the above screening criteria, eligible CHD patients and CHD-DM2 patients were selected for further analysis and study. Missing data imputation When the proportion of missing values in the data exceeds 20%, the data will be excluded from the final analysis dataset to ensure data integrity and analytical accuracy.。For cases where the proportion of missing values is below 20%, we will use the random forest regression method for imputation to effectively estimate and fill in the missing values, thereby maintaining the integrity of the dataset and the continuity of the analysis. Data imbalance handling In clinical data research, missing data can lead to a decline in model accuracy and even produce incorrect results. Additionally, due to the objective differences in the incidence rates of various diseases, the number of positive and negative samples is often highly imbalanced, making the class imbalance problem very common in clinical medical data, and leading to poor classification accuracy for minority class samples in the model [ 11 – 12 ] . To address the data imbalance problem, this study used the SMOTENC algorithm, combined with the themis package in R (version 3.6.1) for data preprocessing. The SMOTENC algorithm is used to generate synthetic samples for the minority class to balance the class distribution in the dataset. Feature factor selection Feature selection is an important and commonly used dimensionality reduction technique, which obtains the optimal feature subset by removing irrelevant and redundant information from the dataset [ 13 – 14 ] .It is also a knowledge discovery tool that provides in-depth insights into the problem by explaining the most relevant features [ 15 ] . The Lasso regression algorithm achieves dimensionality reduction and variable selection for high-dimensional data [ 16 ] . In this study, feature selection combined univariate analysis with Lasso regression.First, significant variables were selected through univariate analysis to initially identify potential candidate features. Then, the Lasso regression method was applied to further select these candidate features, introducing a penalty term to reduce model complexity and ultimately determine the most predictive feature set. Model construction Machine learning methods are becoming increasingly popular in medical research [17] .Supervised classification techniques are popular machine learning methods that aim to explain the dependent variable based on independent variables [ 18 ] . SVM was originally introduced by Cortes and Vapnik and is now a popular classification tool in machine learning, statistics, and pattern recognition [19–20] . The K-nearest neighbor algorithm (kNN) is a supervised machine learning algorithm mainly used for classification and prediction purposes [ 21 ] .Extreme Gradient Boosting (XGBoost) is a simple yet effective machine learning model, consisting of a combination of decision tree learning and gradient boosting [22–23] . We divided the included patients into a training set and a testing set in a 7:3 ratio. This study used five machine learning models (Logistic regression, Logistic_Lasso, KNN, SVM, and XGBoost) to construct prediction models. The Logistic regression, SVM, KNN, and XGBoost models were implemented using the “Caret” package in R software, while the Logistic_Lasso model was implemented using the “Caret” package combined with the “glmnet” package.The selection of model hyperparameters was performed using ten-fold cross-validation on the training dataset. Cross-validation ensures better evaluation of the model’s performance by averaging the metrics from multiple trials. The model's predictive ability was validated using the confusion matrix and the area under the receiver operating characteristic (ROC) curve (AUC).The clinical utility of the model was evaluated using the clinical decision curve analysis (DCA). Variable importance assessment In machine learning algorithms, feature importance refers to the degree to which a feature variable affects the target variable. The selection of features significantly impacts the predictive accuracy of the algorithm. An excessive or insufficient number of features can lead to overfitting or underfitting, respectively, preventing the model from achieving optimal accuracy [ 24 ] . Perform feature importance analysis on the selected model to determine the impact of each feature on the prediction. The importance of each feature is calculated based on the internal mechanisms of their respective models. The "Caret" package in R software is used to perform a comprehensive analysis and comparison of variable importance across different models.Additionally, we evaluated and validated the results using the optimal model. Shapley Additive explanations (SHAP) model SHAP is a game-theoretic technique used to explain the output of any machine learning model [ 25 ] . In recent years, SHAP has been proposed as one of the methods for interpreting machine learning and even deep learning models. Its functions include describing the overall contribution of features, explaining their specific impact on various samples, examining feature interactions, and analyzing the joint effects of feature dependencies [ 26 ] . After selecting the best model, the SHAP model is used to display the relationships between the importance of each feature and analyze and interpret the machine learning model results. In R software, the contribution value of each variable included in the pre-designed model is displayed using histograms of varying lengths. This visualization allows for an unbiased evaluation of each variable's contribution within the system, meaning that the impact of each variable's value on the model can be independently considered and assessed [27] . Statistical analysis The Shapiro-Wilk test is used to assess the normality of continuous variable distributions, and the Bartlett test is used to assess the homogeneity of variance in continuous variables. Continuous variables with a normal distribution are presented as mean ± standard deviation (SD), while continuous variables with a non-normal distribution are represented by median and interquartile range (IQR). For variables that follow a normal distribution and have homogeneous variances, a t-test is used for comparison; otherwise, the Mann-Whitney U test is used for comparison. Categorical variables are expressed as percentages and analyzed using the chi-square test or Fisher's exact test. A two-sided p-value less than 0.05 is considered statistically significant. All statistical analyses are performed using R (version 3.6.1). Results Baseline analysis A total of 29,960 cardiovascular disease patients were screened according to strict inclusion and exclusion criteria, excluding 2,681 non-CHD patients and 19,657 samples with missing key clinical information. Baseline data analysis was conducted on the remaining 5,681 eligible patients. There were 4,396 patients in the CHD group, accounting for 77.3% of the sample, and 1,285 patients in the CHD-DM2 group, accounting for 22.6%. Detailed results are shown in Table 1 . In this study, we performed a comprehensive and systematic comparative analysis of the baseline characteristics between the CHD-DM2 group and the CHD group.The results show that there are 52 indicators with differences in baseline information between the two groups.Specifically, in terms of age distribution, patients in the CHD-DM2 group were significantly younger than those in the CHD group (61 years vs. 63 years, P < 0.01).The weight of patients in the CHD-DM2 group was also lower than that in the CHD group (74kg [65.0;82.0] vs. 75kg [66.0;83.0], P < 0.05). The occupational distribution shows that the proportion of intellectual workers in the CHD-DM2 group is higher (26.0%), significantly greater than the 22.5% in the CHD group (P < 0.01), which may be related to factors such as lifestyle, dietary habits, and work stress in specific occupational groups. In terms of hypertension prevalence, the CHD-DM2 group was significantly lower than the CHD group (48.9% vs. 62.5%, P < 0.001). In addition, the pulse rate in the CHD-DM2 group was lower than in the CHD group (76 times/min vs. 78 times/min, P < 0.001).Further analysis revealed that there were also significant differences between the CHD-DM2 and CHD groups in indicators such as white blood cell (WBC) count, red blood cell (RBC) count, hemoglobin (HB) content, mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) (P < 0.05).Notably, the CHD-DM2 group had lower mean platelet volume (MPV) and platelet distribution width (PDW) compared to the CHD group (P < 0.001), suggesting that platelet function or activity may be regulated by the diabetic state.In addition, the CHD-DM2 group had a lower positive rate for proteinuria (Pro) and glycosuria (Glu) compared to the CHD group (4.91% vs. 8.64%; 5.14% vs. 26.3%; P < 0.001), further highlighting the specific effects of type 2 diabetes on kidney function and glucose metabolism.In terms of coagulation function, the CHD-DM2 group had a longer prothrombin time (PT) (11.3s vs. 11.1s, P < 0.001), while the prothrombin time activity percentage (PT.Activity) was slightly lower (102% vs. 105%, P < 0.001), which may be related to an imbalance in the coagulation and fibrinolytic systems associated with diabetes.Biochemically, the CHD-DM2 group had slightly lower blood urea nitrogen (BUN) levels (5.30mmol/L vs. 5.40mmol/L, P < 0.001), while their uric acid (Ua) levels were higher (317µmol/L vs. 304µmol/L, P < 0.001), suggesting differences in metabolic states. Finally, the blood glucose (BG) levels in the CHD-DM2 group were significantly lower than those in the CHD group (5.01mmol/L vs. 6.79mmol/L, P < 0.001), directly reflecting the diagnostic features of diabetes and its stringent requirements for blood glucose management.Detailed information is provided in Table 1 . Table 1 Baseline characteristics of study subjects ALL N = 5681 CHD-DM2 N = 4396 CHD N = 1285 p.overall Gender, no. (%) 0.069 Male 4004 (70.5%) 3125 (71.1%) 879 (68.4%) Female 1677 (29.5%) 1271 (28.9%) 406 (31.6%) Age (years), median (IQR) 61.0 [54.0;70.0] 61.0 [53.0;70.0] 63.0 [55.0;71.0] 0.001 Height (cm), median (IQR) 170 [162;174] 170 [162;174] 170 [162;174] 0.429 Weight (kg), median (IQR) 74.0 [65.0;83.0] 74.0 [65.0;82.0] 75.0 [66.0;83.0] 0.041 National, no. (%) 0.009 han nationality 3641 (64.1%) 2802 (63.7%) 839 (65.3%) uyghur 1377 (24.2%) 1057 (24.0%) 320 (24.9%) kazakh 260 (4.58%) 225 (5.12%) 35 (2.72%) hui ethnic group 280 (4.93%) 215 (4.89%) 65 (5.06%) mongolian 60 (1.06%) 44 (1.00%) 16 (1.25%) other nationalities 63 (1.11%) 53 (1.21%) 10 (0.78%) Educational.Level, no. (%) 0.924 primary school 1196 (21.1%) 928 (21.1%) 268 (20.9%) junior middle school 1208 (21.3%) 938 (21.3%) 270 (21.0%) senior middle school 1654 (29.1%) 1270 (28.9%) 384 (29.9%) university 1623 (28.6%) 1260 (28.7%) 363 (28.2%) Professional, no. (%) 0.004 Professional and technical personnel in institute 1434 (25.2%) 1145 (26.0%) 289 (22.5%) merchant and pasture 1371 (24.1%) 1076 (24.5%) 295 (23.0%) retirees and other personnel 2876 (50.6%) 2175 (49.5%) 701 (54.6%) Current.Smoker, no. (%) 0.133 non-smoking 3263 (57.4%) 2501 (56.9%) 762 (59.3%) smoking 2418 (42.6%) 1895 (43.1%) 523 (40.7%) Current.Drinker, no. (%) 0.149 non-drinking 4460 (78.5%) 3432 (78.1%) 1028 (80.0%) drinking 1221 (21.5%) 964 (21.9%) 257 (20.0%) Hypertensive.history, no. (%) < 0.001 No 2729 (48.0%) 2247 (51.1%) 482 (37.5%) Yes 2952 (52.0%) 2149 (48.9%) 803 (62.5%) Diabetes.History. (%) 0 No 4439 (78.1%) 4055 (92.2%) 384 (29.9%) Yes 1242 (21.9%) 341 (7.76%) 901 (70.1%) Arrhythmia, no. (%) 0.213 No 5611 (98.8%) 4337 (98.7%) 1274 (99.1%) Yes 70 (1.23%) 59 (1.34%) 11 (0.86%) Hyperlipidemia, no. (%) 0.749 No 5635 (99.2%) 4359 (99.2%) 1276 (99.3%) Yes 46 (0.81%) 37 (0.84%) 9 (0.70%) Cerebral.Infarction, no. (%) 0.342 No 5569 (98.0%) 4314 (98.1%) 1255 (97.7%) Yes 112 (1.97%) 82 (1.87%) 30 (2.33%) Angina, no. (%) 0.983 No 4966 (87.4%) 3842 (87.4%) 1124 (87.5%) Yes 715 (12.6%) 554 (12.6%) 161 (12.5%) Heart.failure, no. (%) 1 No 5664 (99.7%) 4383 (99.7%) 1281 (99.7%) Yes 17 (0.30%) 13 (0.30%) 4 (0.31%) Family.history, no. (%) 0.232 No 5144 (90.5%) 3992 (90.8%) 1152 (89.6%) Yes 537 (9.45%) 404 (9.19%) 133 (10.4%) Pulse (times/min), median (IQR) 76.0 [69.0;83.0] 76.0 [69.0;82.0] 78.0 [70.0;85.0] < 0.001 WBC (×10 9 /L), median (IQR) 6.62 [5.54;7.95] 6.54 [5.48;7.90] 6.84 [5.72;8.10] < 0.001 NE%, median (IQR) 58.4 [52.8;64.7] 58.3 [52.6;64.7] 59.0 [53.4;64.7] 0.062 LY%, median (IQR) 31.3 [25.4;36.9] 31.4 [25.2;37.0] 30.8 [25.5;36.5] 0.135 MO%, median (IQR) 6.60 [5.40;7.90] 6.60 [5.40;8.00] 6.50 [5.40;7.90] 0.372 EO%, median (IQR) 2.10 [1.30;3.30] 2.10 [1.30;3.40] 2.10 [1.30;3.30] 0.65 BA%, median (IQR) 0.30 [0.20;0.50] 0.30 [0.20;0.50] 0.30 [0.20;0.50] 0.101 NE1 (×10 9 /L), median (IQR) 3.84 [3.05;4.85] 3.78 [3.01;4.84] 3.99 [3.21;4.95] < 0.001 LY1 (×10 9 /L), median (IQR) 2.01 [1.59;2.49] 2.00 [1.58;2.48] 2.07 [1.65;2.53] 0.004 MO1 (×10 9 /L), median (IQR) 0.44 [0.34;0.56] 0.43 [0.34;0.56] 0.45 [0.35;0.58] 0.005 EO1 (×10 9 /L), median (IQR) 0.14 [0.08;0.23] 0.14 [0.08;0.23] 0.14 [0.08;0.23] 0.39 BA1 (×10 9 /L), median (IQR) 0.02 [0.01;0.04] 0.02 [0.01;0.04] 0.02 [0.01;0.04] 0.661 RBC (×10 12 /L), median (IQR) 4.64 [4.29;4.99] 4.64 [4.30;5.00] 4.61 [4.26;4.97] 0.035 HB (g/L), median (IQR) 141 [130;151] 141 [131;151] 139 [129;149] < 0.001 HCT (L/L), median (IQR) 0.48 [0.42;41.8] 0.48 [0.42;41.8] 31.9 [0.42;41.7] 0.13 MCV (fL), median (IQR) 91.3 [88.5;94.2] 91.5 [88.6;94.4] 90.6 [87.9;93.6] < 0.001 MCH (pg), median (IQR) 30.3 [29.2;31.4] 30.3 [29.3;31.4] 30.1 [29.1;31.2] < 0.001 MCHC (g/L), median (IQR) 331 [323;338] 331 [323;338] 331 [324;339] 0.066 RDW (%), median (IQR) 13.0 [12.6;13.5] 13.0 [12.6;13.5] 13.0 [12.6;13.5] 0.641 PLT (×10 9 /L), median (IQR) 219 [183;260] 219 [184;261] 217 [182;258] 0.262 MPV (fL), median (IQR) 10.7 [10.0;11.4] 10.6 [10.0;11.4] 10.8 [10.2;11.5] < 0.001 PCT (%), median (IQR) 0.23 [0.20;0.28] 0.23 [0.20;0.28] 0.24 [0.20;0.28] 0.606 PDW (fL), median (IQR) 16.1 [14.6;16.9] 16.1 [14.5;16.9] 16.2 [15.0;17.1] < 0.001 Pro, no. (%) < 0.001 negative 5354 (94.2%) 4180 (95.1%) 1174 (91.4%) positive 327 (5.76%) 216 (4.91%) 111 (8.64%) ERY, no. (%) 0.502 negative 5012 (88.2%) 3871 (88.1%) 1141 (88.8%) positive 669 (11.8%) 525 (11.9%) 144 (11.2%) Glu, no. (%) < 0.001 negative 5117 (90.1%) 4170 (94.9%) 947 (73.7%) positive 564 (9.93%) 226 (5.14%) 338 (26.3%) PT (s), median (IQR) 11.2 [10.7;11.8] 11.3 [10.8;11.8] 11.1 [10.6;11.8] < 0.001 PT Activity (%), median (IQR) 102 [91.7;117] 102 [91.3;115] 105 [92.0;122] < 0.001 ISR, median (IQR) 1.02 [0.96;1.20] 1.02 [0.96;1.20] 1.01 [0.95;1.17] 0.002 NC, median (IQR) 11.2 [10.5;11.5] 11.2 [10.5;11.5] 11.3 [10.6;11.5] 0.219 FIB (g/L), median (IQR) 3.38 [2.95;3.80] 3.36 [2.94;3.78] 3.44 [3.00;3.83] 0.004 PPT (s), median (IQR) 31.0 [29.1;33.1] 31.0 [29.2;33.2] 30.7 [28.9;32.9] 0.002 APTT-R(s), median (IQR) 1.00 [0.90;1.10] 1.00 [0.90;1.10] 1.00 [0.90;1.10] 0.01 TT Verus (s), median (IQR) 20.2 [19.2;21.3] 20.2 [19.2;21.3] 20.2 [19.2;21.3] 0.534 APTT (s), median (IQR) 30.8 [30.2;31.3] 30.8 [30.2;31.4] 30.7 [30.2;31.2] 0.032 TT (s), median (IQR) 20.3 [20.0;20.7] 20.3 [20.0;20.7] 20.3 [20.0;20.7] 0.632 BUN (mmol/L), median (IQR) 5.30 [4.40;6.38] 5.30 [4.30;6.30] 5.40 [4.50;6.56] < 0.001 Cr (µmol/L), median (IQR) 71.0 [61.0;82.7] 71.2 [61.0;82.2] 71.0 [60.0;83.0] 0.714 Ua (µmol/L), median (IQR) 314 [260;374] 317 [263;377] 304 [252;366] < 0.001 BG (mmol/L), median (IQR) 5.21 [4.64;6.57] 5.01 [4.56;5.85] 6.79 [5.38;9.29] < 0.001 HbA1c (mmol/L), median (IQR) 2.21 [2.01;2.48] 2.16 [1.98;2.39] 2.45 [2.17;2.80] < 0.001 TG (mmol/L), median (IQR) 1.50 [1.08;2.14] 1.48 [1.06;2.09] 1.60 [1.15;2.36] < 0.001 TC (mmol/L), median (IQR) 3.80 [3.10;4.56] 3.83 [3.14;4.56] 3.68 [2.98;4.55] < 0.001 HDL (mmol/L), median (IQR) 1.02 [0.85;1.23] 1.03 [0.87;1.24] 0.98 [0.81;1.18] < 0.001 LDL (mmol/L), median (IQR) 2.39 [1.84;3.07] 2.42 [1.87;3.09] 2.28 [1.73;3.01] < 0.001 APO.AI (g/L), median (IQR) 1.13 [1.00;1.29] 1.14 [1.01;1.30] 1.12 [0.98;1.26] < 0.001 APO.B (g/L), median (IQR) 0.79 [0.62;0.98] 0.79 [0.63;0.98] 0.77 [0.60;0.96] 0.019 LP (mg/L), median (IQR) 153 [85.3;278] 154 [85.0;276] 150 [86.6;292] 0.988 STB (µmol/L), median (IQR) 11.0 [8.30;14.9] 11.2 [8.40;15.0] 10.6 [8.00;14.2] 0.002 DBIL (µmol/L), median (IQR) 3.39 [2.30;4.80] 3.40 [2.31;4.84] 3.31 [2.24;4.63] 0.068 IBIL (µmol/L), median (IQR) 7.66 [5.10;10.8] 7.80 [5.21;11.0] 7.23 [4.83;10.4] 0.001 TP (g/L), median (IQR) 66.1 [62.7;70.2] 66.1 [62.6;70.2] 66.1 [62.8;70.2] 0.373 Glb (g/L), median (IQR) 25.8 [22.8;29.1] 25.8 [22.8;29.2] 25.9 [22.7;29.1] 0.963 AS (mm), median (IQR) 33.0 [30.0;35.0] 33.0 [30.0;35.0] 32.0 [30.0;35.0] 0.128 AVA (mm), median (IQR) 20.0 [19.0;22.0] 20.0 [19.0;22.0] 20.0 [19.0;22.0] 0.92 LA (mm), median (IQR) 34.0 [31.0;36.0] 33.0 [31.0;36.0] 34.0 [31.0;37.0] 0.001 LVED (mm), median (IQR) 49.0 [46.0;51.0] 48.0 [46.0;51.0] 49.0 [46.0;52.0] 0.042 LVES (mm), median (IQR) 32.0 [30.0;34.0] 32.0 [30.0;34.0] 32.0 [30.0;34.0] 0.024 IVS (mm), median (IQR) 9.00 [9.00;10.0] 9.00 [9.00;10.0] 9.00 [9.00;10.0] 0.675 LVPW (mm), median (IQR) 9.00 [9.00;10.0] 9.00 [9.00;10.0] 9.00 [9.00;10.0] 0.274 LVOT (mm), median (IQR) 28.0 [26.0;29.0] 27.5 [26.0;29.0] 28.0 [26.0;29.0] 0.781 RV (mm), median (IQR) 18.0 [18.0;20.0] 18.0 [18.0;20.0] 18.0 [17.0;20.0] 0.732 RA (mm), median (IQR) 33.0 [31.0;35.0] 33.0 [31.0;35.0] 33.0 [31.0;35.0] 0.312 PA (mm), median (IQR) 23.0 [21.0;24.0] 23.0 [21.0;24.0] 23.0 [21.0;25.0] 0.207 MVE (m/s), median (IQR) 0.71 [0.61;0.83] 0.72 [0.61;0.84] 0.70 [0.60;0.82] 0.052 MVA (L/min), median (IQR) 0.73 [0.62;0.86] 0.72 [0.62;0.85] 0.76 [0.65;0.91] < 0.001 MVEA, median (IQR) 0.94 [0.75;1.25] 1.01 [0.76;1.26] 0.85 [0.72;1.21] < 0.001 FS (%), median (IQR) 34.0 [32.0;36.0] 34.0 [33.0;36.0] 34.0 [32.0;36.0] 0.001 EF (%), median (IQR) 63.0 [60.0;65.0] 63.0 [60.0;65.0] 62.0 [59.0;65.0] 0.037 SV (%), median (IQR) 70.0 [63.0;78.0] 70.0 [63.0;78.0] 71.0 [63.0;78.0] 0.606 CO (L/min), median (IQR) 5.11 [4.46;5.82] 5.11 [4.45;5.81] 5.15 [4.51;5.86] 0.108 LTS, no. (%) 0.006 No 5273 (92.8%) 4103 (93.3%) 1170 (91.1%) Yes 408 (7.18%) 293 (6.67%) 115 (8.95%) LAD, no. (%) < 0.001 No 2166 (38.1%) 1793 (40.8%) 373 (29.0%) Yes 3515 (61.9%) 2603 (59.2%) 912 (71.0%) D1.2 Narrow, no. (%) < 0.001 No 4424 (77.9%) 3477 (79.1%) 947 (73.7%) Yes 1257 (22.1%) 919 (20.9%) 338 (26.3%) LCX, no. (%) < 0.001 No 3373 (59.4%) 2749 (62.5%) 624 (48.6%) Yes 2308 (40.6%) 1647 (37.5%) 661 (51.4%) OM, no. (%) < 0.001 No 4939 (86.9%) 3894 (88.6%) 1045 (81.3%) Yes 742 (13.1%) 502 (11.4%) 240 (18.7%) RCA, no. (%) < 0.001 No 3052 (53.7%) 2480 (56.4%) 572 (44.5%) Yes 2629 (46.3%) 1916 (43.6%) 713 (55.5%) Note: Diabetes.History, family history of diabetes (father, mother, and/or siblings); Hypertensive.history, family history of hypertension (father, mother, and/or siblings); WBC, white blood cell; NE, neutrophil; LY, lymphocyte; MO, monocyte; EO, eosinophil; BA, basophil; NE1, neutrophil count; LY1, lymphocyte count; MO1, monocyte count; EO1, eosinophil count; BA1, basophil count; RBC, red blood cell; HB, hemoglobin; HCT, hematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; RDW, red blood cell distribution width; PLT, platelet count; MPV, mean platelet volume; PCT, platelet volume ratio; PDW, platelet distribution width; Pro, proteinuria; ERY, hematuria; Glu, glycosuria; PT, prothrombin time; PT.Activity, prothrombin time (PT) activity percentage; ISR, international standardized ratio; NC, normal control; FIB, fibrinogen; PPT, partial prothrombin time; APTT-R, activated partial thromboplastin time ratio; TT Verus, thrombin time; APTT, APTT control; TT, TT normal control; BUN, blood urea nitrogen; Cr, creatinine; Ua, uric acid; BG, blood glucose; HbA1c, glycated hemoglobin; TG, triglyceride; TC, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein; APO.AI, apolipoprotein AI; APO.B, apolipoprotein B; LP, lipoprotein; STB, total bilirubin; DBIL, direct bilirubin; IBIL, indirect bilirubin; TP, total protein; Glb, serum globulin; AS, aortic valve stenosis; AVA, valve area; LA, left atrial diameter; LVED, left ventricular end-diastolic diameter (Left Ventricular End Diastolic Diameter); LVES, left ventricular end-systolic dimension (Left Ventricular End Systolic Dimension); IVS, interventricular septal thickness; LVPW, left ventricular posterior wall thickness; LVOT, left ventricular outflow tract (Left ventricular outflow tract); RV, right ventricular diameter; RA, right atrial diameter; PA, pulmonary artery diameter; MVE, mitral valve E-wave velocity (Mitral Valve E-wave velocity); MVA, mitral valve area (Maximum Ventricular Ejection per Minute); MVEA, mitral valve E/A ratio (Mitral Valve E/A Ratio); FS, fractional shortening (Fractional Shortening); EF, left ventricular ejection fraction (Ejection Fraction); SV, stroke volume (Stroke Volume); CO, cardiac output (Cardiac Output); LTS, left main coronary artery stenosis; LAD, left anterior descending coronary artery; D1.2.Narrow, diagonal branch 1 and 2 stenosis; LCX, left circumflex coronary artery; RCA, right coronary artery Imbalance Handling Table 2 shows a comparison between the original dataset and the balanced dataset obtained using the SMOTENC algorithm. As shown in Table 2 , the original dataset exhibits significant bias with an imbalance rate of 3.421. To address this imbalance and prevent result distortion, we used the smotenc function from the themis package in R (version 3.6.1) to balance the data. Based on statistical considerations, the value of k is set to 10, and different k values are tested for each dataset. The parameter is set to over_ratio = 1, which represents the ratio of majority to minority frequencies. We performed oversampling on the category with the fewest observations and undersampling on the category with the most observations, thereby balancing the CHD patient group (1285) and the CHD-DM2 patient group (4396). Table 2 and Fig. 1 provide detailed distributions of observations for each category in the balanced and imbalanced training datasets. Table 2 Description of the original and balanced data. Dataset Minority class Majority class Samples in total Imbalance rate CHD(original) 1285 4396 5681 3.421 CHD(SMOTEN) 4396 4396 8792 1 Feature Selection We adopted an integrated strategy that cleverly combines univariate analysis with Lasso regression, achieving efficient and precise feature selection. First, we used univariate analysis as a preliminary screening method, effectively narrowing down the initial set of variables to the most promising 52. During this process, we strictly controlled the error rate to ensure the robustness of the selection. Then, to further refine the feature set and enhance the model's predictive performance, we introduced Lasso regression. In the Lasso model, we implemented a strict ten-fold cross-validation strategy to comprehensively evaluate the model's performance on different data subsets. Using ten-fold cross-validation to select features with non-zero regression coefficients (λ), we successfully reduced the initially selected 52 variables to 25 key variables.The variables Age, Diabetes.History, Hypertensive.history, Weight, Pulse, LY1, MCH, PDW, Pro, Glu, PT, PT.Activity, PPT, APTT, BG, TG, APO.B, IBIL, LVES, MVA, FS, EF, LAD, LCX, and OM were identified as the final predictive factors with the greatest impact on the model's performance (Fig. 2 A, B). Model Construction and Evaluation The study population (n = 5681) was randomly divided into a training set (n = 3977) and a testing set (n = 1704) at a 7:3 ratio. The training set was used to develop the predictive model, while the testing set was used for further validation. Participants were clearly classified into the CHD group (n = 4396) and the CHD-DM2 group (n = 1285), with the predictive model based on whether participants had CHD-DM2. The overall study workflow is shown in Fig. 3 , which clearly outlines the steps of data partitioning, model construction, and validation. Using the training and testing sets, we constructed and validated five common machine learning models: Logistic, Logistic_Lasso, SVM, kNN, and XGBoost. When training the models on the balanced dataset, the accuracy, sensitivity, and specificity of all five models were significantly improved. As shown in Table 3 , the XGBoost model performed particularly well on the balanced dataset. Its AUC value increased from 87.06% (95% CI: 84.85%-88.15%) on the imbalanced dataset to 95.94% (95% CI: 89.42%-91.6%), accuracy improved from 0.8656 to 0.9594, sensitivity increased from 0.7661 to 0.9054, and specificity improved from 0.8761 to 0.9150, showing superior performance compared to the other models. Figure 4 shows the ROC curves of five machine learning models trained on both the balanced and imbalanced datasets. All models performed better when trained on the balanced dataset. The AUC of the XGBoost model on the imbalanced dataset reached 0.8706, significantly higher than the other four models (Fig. 4 A). On the balanced dataset, the AUC of the XGBoost model increased to 0.9594, again outperforming the other four models (Fig. 4 B).The DCA curve illustrates the clinical utility of the predictive model by evaluating the net benefit at different threshold probabilities [ 28] .The DCA curve (Fig. 5 ) reveals the high clinical value of the XGBoost model, which achieves the highest net benefit at the lowest misdiagnosis rate. Overall, these findings indicate that the XGBoost model significantly outperforms the other four machine learning models, showing the most stable performance. Finally, Fig. 6 visually illustrates the positive impact of balancing on the dataset and model performance through a radar chart. The errors of all five machine learning models are significantly lower on the balanced dataset compared to the imbalanced dataset, further emphasizing the critical role of data balance in improving model performance. Table 3 Performance metrics of five machine learning models on the original and balanced datasets Model AUC Accuracy Sensitivity Specificity PPV NPV Precision (%) Recall (%) Imbalanced Data Model KNN 0.5957 0.618 0.5242 0.6441 0.2915 0.8290 0.5241 0.2914 SVM 0.8500 0.882 0.6828 0.9377 0.7537 0.9137 0.7581 0.5912 XGBoost 0.8706 0.8656 0.7661 0.8934 0.6674 0.9319 0.7849 0.6213 Logistic_Lasso 0.8500 0.8539 0.7204 0.8761 0.6358 0.9225 0.7742 0.6357 Logistic 0.8323 0.8791 0.7204 0.9234 0.7243 0.9220 0.7661 0.6063 Balanced Data Model KNN 0.6811 0.6331 0.7994 0.4662 0.6003 0.6985 0.5466 0.6667 SVM 0.9103 0.8324 0.7426 0.9226 0.9508 0.7814 0.8213 0.8353 XGBoost 0.9594 0.9102 0.9054 0.9150 0.9144 0.9060 0.9054 0.9144 Logistic_Lasso 0.8609 0.8241 0.7517 0.8922 0.8749 0.7818 0.7517 0.8749 Logistic 0.8523 0.8237 0.7388 0.9233 0.9045 0.7696 0.7388 0.8722 Variable Importance Analysis Figure 7 illustrates the key contributions of various variables to prediction in the five predictive models trained on the balanced dataset. In the kNN model, Diabetes History, blood glucose levels, HbA1c levels, positive urine glucose test, mitral valve area, and left anterior descending artery were identified as the most important predictive factors. Similarly, for the SVM model, Diabetes History, blood glucose levels, HbA1c levels, positive urine glucose test, mitral valve area, left circumflex artery, and family history of hypertension were also considered key predictive factors. In the XGBoost model, Diabetes History, APTT ratio, International Normalized Ratio (INR), blood glucose levels, race, APTT control, and HbA1c levels are key predictive factors.On the one hand, the Logistic regression model emphasizes the high importance of Diabetes History, positive urine glucose test, Platelet Distribution Width, indirect bilirubin, obtuse margin branch, and obtuse margin branch. Furthermore, the Logistic_Lasso model emphasizes the importance of Diabetes History, positive urine glucose test, obtuse margin branch, mitral valve area level, left anterior descending coronary artery stenosis, and hypertensive family history. In conclusion, the four variables—family history of diabetes, blood glucose level, positive urine glucose test, and glycated hemoglobin—are important in different models. This highlights their importance in influencing CHD-DM2 risk, indicating that these factors are key targets for risk assessment and potential intervention strategies. Model interpretability To more intuitively illustrate the role of the selected variables in predicting CHD-DM2, we used SHAP interpretability analysis, as shown in the summary plot (Fig. 8 A).In this plot, each feature importance line represents the impact of different variables on the model’s prediction. Each point corresponds to the contribution of a single patient to the prediction of that specific variable, with purple points indicating high feature values and yellow points indicating low feature values. This study identifies variables independently associated with CHD-DM2, including diabetes history, blood glucose levels, prothrombin time, glycated hemoglobin, APTT control, age, indirect bilirubin, the activity percentage of prothrombin time (PT), left atrial diameter, and platelet distribution width (Fig. 8 A). Figure 8 B shows the top ten features ranked by importance. Clearly, diabetes history, blood glucose levels, and prothrombin time are the three most critical factors driving the prediction results, while glycated hemoglobin, APTT control, age, indirect bilirubin, and the activity percentage of prothrombin time (PT) are important supplementary factors. Discussion Over the past decade, artificial intelligence (AI),machine learning and deep learning has become a popular subject both within and outside of the scientific community [29] .Machine learning can use sample data to build models for predicting future outcomes or identifying hidden underlying patterns in the input data [ 30 ] . The expansion of biological data size and the increasing complexity have led to the growing application of machine learning in the field of biology [31] . Artificial intelligence (AI) methods have great potential to surpass traditional and domain-specific approaches in prediction accuracy, operational efficiency, and other aspects [ 32 ] . Machine learning, as a powerful set of algorithms, utilizes precise identification of complex patterns in data to perform key tasks such as prediction and classification. With the increasing richness of data resources and the leap in computational power, machine learning technology has been widely applied and deeply integrated in both industry and natural sciences, demonstrating its irreplaceable value and potential [ 33 – 34 ] . This study uses the SMOTENC algorithm combined with advanced machine learning techniques to handle imbalanced data, thereby minimizing potential bias in the results [35] . By comparing the performance of five machine learning models (kNN, SVM, XGBoost, Logistic_Lasso, Logistic) before and after balancing, the results show that the data balancing strategy significantly improved the overall performance of the models, particularly the XGBoost model, which exhibited the best performance on the balanced dataset, achieving the highest precision, recall, and AUC values. The onset and progression of coronary heart disease (CHD) is a complex process influenced by both genetic and environmental factors, involving the combined effects of numerous variables. Extensive previous research has identified a range of key risk factors for CHD, including but not limited to age, gender, hypertension, diabetes, smoking habits, and dyslipidemia [ 36 ] . To further analyze how these selected factors specifically impact the prediction model for CHD-DM2, we used SHAP to interpret the model. In explaining the model's predictions, SHAP values provide valuable information that helps us understand which features are most important in predicting coronary heart disease. This approach not only enhances the transparency of the model but also offers direction for future research and interventions [ 37 ] . Our study indicates that a history of diabetes, blood glucose levels, positive urine glucose test, and HbA1c are major risk factors for CHD-DM2. Given these factors, we must focus on reducing the risk of cardiovascular diseases in patients with type 2 diabetes [ 38 ] . Type 2 diabetes is a metabolic disorder characterized by chronic hyperglycemia and improper lipid, carbohydrate, and protein metabolism caused by insulin resistance and insufficient insulin secretion, leading to persistent increases in blood pressure, blood lipids, and blood glucose, thus disrupting normal body functions and metabolic balance. Epidemiological evidence indicates that any increase in blood glucose, even within the prediabetic range, increases the risk of cardiovascular diseases [ 39 ] . In response to this situation, it is crucial to emphasize the importance of controlling blood glucose and lipid levels in CHD-DM2 patients. Regular monitoring and maintaining blood glucose within the ideal range, along with reducing the intake of high-sugar and high-fat foods, can effectively alleviate the irreversible damage caused by long-term hyperlipidemia and hyperglycemia, thereby improving patients' quality of life and delaying disease progression. In recent years, scholars and experts both domestically and internationally have generally believed that the incidence of coronary heart disease is closely related to hypertension and hyperglycemia, and based on this understanding, they have actively conducted research on treatment methods for CHD-DM2. The progress in this field urgently requires high clinical attention and should be put into practice. Specifically, when conducting clinical treatment for CHD-DM2 patients, it is crucial to monitor and record the patient's various risk factors and implement targeted interventions. With the continuous improvement of medical conditions and the deepening of related research, we believe that more scientifically effective methods to control and treat CHD-DM2 will emerge in the future, thereby providing better treatment and services for patients. Abbreviations T2DM type 2 diabetes CHD coronary heart disease CHD-DM2 coronary heart disease combined with type 2 diabetes. Declarations Ethics approval and consent to participate The study was approved by the Ethics Committee for Drug Clinical Trials of the First Affiliated Hospital of Xinjiang Medical University(Approval No. XJYKDXR20220725007). Informed consent was obtained from the participants and the study was performed in accrodance with Declaration of Helsinki. No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication.I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously, and not under consideration for publication elsewhere, in whole or in part.All the authors listed have approved the manuscript that is enclosed.That its publication has been approved by the responsible authorities at the institution where the work carried out. Consent for publication The co-authors agreed to publication in “ Scientific Reports ”.The copyright to the article is transferred to “ Scientific Reports ” effective if and when the article is accepted for publication. The author warrants that his/her contribution is original and that he/she has full power to make this grant. The author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. Availability of data and material This study selected clinical data from the First Affiliated Hospital of Xinjiang Medical University. If anyone would like to obtain data from this study, please contact Dandan Tang. Competing interests All authors disclosed no relevant relationships. Funding The present paper was supported by the following funding: (1)Youth Science Fund of the Natural Science Foundation of Xinjiang Uyghur Autonomous Region；Project Number: 2022D01C718；Project Title:"Developing a Risk Prediction Model for Concurrent Diabetes in Coronary Heart Disease Patients Using Multimodal Data and Machine Learning".(2)Special Funds for Talents of Xinjiang Medical University；Project Number: 0103010211. Authors' contributions Dandan Tang was responsible for conceptualization, methodology, research design, data analysis, and article writing; Fengwei Liang was responsible for clinical data curation, algorithm development, SHAP implementation, statistical validation; Xingli Gu was responsible for data compilation and statistical analysis; Yuanyuan Jin was responsible for research coordination and material support; Xuanjie Hu was responsible for data collection and compilation; Fen Liu was responsible for clinical interpretation, biomarker analysis, revealing research results; Yining Yang was responsible for study design, funding acquisition, manuscript supervision. References Xu, H. et al. Establishment of a diagnostic model of coronary heart disease in elderly patients with diabetes mellitus based on machine learning algorithms. J. Geriatr. Cardiol. 19 (6), 445–455 (2022). Bähler, C., Huber, C. A., Brüngger, B. & Reich, O. Multimorbidity, health care utilization and costs in an elderly community-dwelling population: a claims data based observational study. BMC Health Serv. Res. 15 , 23 (2015). Sagris, M. et al. Inflammation in Coronary Microvascular Dysfunction. Int. J. Mol. Sci. 22 (24), 13471 (2021). Stone, P. H., Libby, P. & Boden, W. E. Fundamental Pathobiology of Coronary Atherosclerosis and Clinical Implications for Chronic Ischemic Heart Disease Management-The Plaque Hypothesis: A Narrative Review. JAMA Cardiol. 8 (2), 192–201 (2023). Turin, T. C. et al. Diabetes and lifetime risk of coronary heart disease. Prim. Care Diabetes . 11 (5), 461–466 (2017). Aronson, D. & Edelman, E. R. Coronary artery disease and diabetes mellitus. Cardiol. Clin. 32 (3), 439–455 (2014). Xu, Q. et al. Prediction of Atrial Fibrillation in Hospitalized Elderly Patients With Coronary Heart Disease and Type 2 Diabetes Mellitus Using Machine Learning: A Multicenter Retrospective Study. Front. Public. Health . 10 , 842104 (2022). Xiao, H. et al. Disease patterns of coronary heart disease and type 2 diabetes harbored distinct and shared genetic architecture. Cardiovasc. Diabetol. 21 (1), 276 (2022). Mirjalili, S. R. et al. An innovative model for predicting coronary heart disease using triglyceride-glucose index: a machine learning-based cohort study. Cardiovasc. Diabetol. 22 (1), 200 (2023). Forrest, I. S. et al. Machine learning-based marker for coronary artery disease: derivation and validation in two longitudinal cohorts. Lancet 401 (10372), 215–225 (2023). Rahman, M. M. & Davis, D. N. Addressing the class imbalance problem in medical datasets[J]. Int. J. Mach. Learn. Comput. 3 (2), 224 (2013). Thabtah, F. et al. Data imbalance in classification: Experimental evaluation[J]. Inf. Sci. 513 , 429–441 (2020). Abdel Majeed, Y., Awadalla, S. S. & Patton, J. L. Regression techniques employing feature selection to predict clinical outcomes in stroke. PLoS One . 13 (10), e0205639 (2018). Wei, H. et al. Environmental chemical exposure dynamics and machine learning-based prediction of diabetes mellitus. Sci. Total Environ. 806 (Pt 2), 150674 (2022). ALIZADEHSANI, R. et al. Machine learning-based coronary artery disease diagnosis: a comprehensive review [J]. Comput. Biol. Med. 111 , 103346 (2019). Shao, L. et al. LASSO-derived nomogram predicting new-onset diabetes mellitus in patients with kidney disease receiving immunosuppressive drugs. J Clin Pharm Ther. ;47(10):1627–1635. CHERKASSKY V.The nature of statistical learning theory [J].IEEE Trans Neural Netw, 1997, 8(6): 1564. (2022). Alghamdi, M. et al. Predicting diabetes mellitus using SMOTE and ensemble machine learning approach: The Henry Ford ExercIse Testing (FIT) project. PLOS ONE . 12 (7), e0179805 (2017). Huang, S. et al. Applications of Support Vector Machine (SVM) Learning in Cancer Genomics. Cancer Genomics Proteom. 15 (1), 41–51 (2018 Jan-Feb). Zhou, S., Sparse & SVM for Sufficient Data Reduction. IEEE Trans. Pattern Anal. Mach. Intell. ; 44 (9):5560–5571. (2022). Harabor, V. et al. Machine Learning Approaches for the Prediction of Hepatitis B and C Seropositivity. Int. J. Environ. Res. Public. Health . 20 (3), 2380 (2023). Liu, F., Yao, J., Liu, C. & Shou, S. Construction and validation of machine learning models for sepsis prediction in patients with acute pancreatitis. BMC Surg. 23 (1), 267 (2023). Myles, A. J., Feudaie, R. N. & Ln, Y. et aL An htroduction to decision tree modeling[J]. Journal of Chemometrics, 18(6): 275–285 (2004). Ma, Y. et al. Explainable machine learning model reveals its decision-making process in identifying patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation. BMC Cardiovasc. Disord . 23 (1), 91 (2023). FAN Z Y, JIANG J M, CHEN, X. et al. Construction and validation of prognostic models in critically Ill patients with sepsis-associated acute kidney injury: interpretable machine learning approach[J]. J. translational Med. 21 (1), 406 (2023). El-Sappagh, S., Alonso, J. M., Islam, S. M. R., Sultan, A. M. & Kwak, K. S. A multilayer multimodal detection and prediction model based on explainable artificial intelligence for Alzheimer's disease. Sci. Rep. 11 (1), 2660 (2021). Mangiagalli, M., Miccolis, I., Maffé, P., Pogliani, E. M. & Corneo, G. Role of granulocyte colony-stimulating factor in relapsed/resistant intermediate and high-grade non-Hodgkin's lymphoma patients treated with the E-SHAP regimen. Tumori. Mar-Apr;81(2):91 – 5. (1995). Kondo, T. et al. Predicting Stroke in Heart Failure and Preserved Ejection Fraction Without Atrial Fibrillation. Circ. Heart Fail. 16 (7), e010377 (2023). Choi, R. Y., Coyner, A. S., Kalpathy-Cramer, J., Chiang, M. F. & Campbell, J. P. Introduction to Machine Learning, Neural Networks, and Deep Learning. Transl Vis. Sci. Technol. 9 (2), 14 (2020). Thapa, R. et al. Predicting Falls in Long-term Care Facilities: Machine Learning Study. JMIR Aging . 5 (2), e35373 (2022). Greener, J. G., Kandathil, S. M., Moffat, L. & Jones, D. T. A guide to machine learning for biologists. Nat. Rev. Mol. Cell. Biol. 23 (1), 40–55 (2022). Martin-Morales, A. et al. Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables. Nutrients 15 (18), 3937 (2023). Zhang, H. et al. PTX3 mediates the infiltration, migration, and inflammation-resolving-polarization of macrophages in glioblastoma. CNS Neurosci. Ther. 28 (11), 1748–1766 (2022). Hu, M. et al. A Risk Prediction Model Based on Machine Learning for Cognitive Impairment Among Chinese Community-Dwelling Elderly People With Normal Cognition: Development and Validation Study. J. Med. Internet Res. 23 (2), e20298 (2021). Thakur, V. S., Kankar, P. K., Parey, A., Jain, A. & Jain, P. K. The implication of oversampling on the effectiveness of force signals in the fault detection of endodontic instruments during RCT. Proc. Inst. Mech. Eng. H . 237 (8), 958–974 (2023). Feitosa, M. F. et al. Heterogeneity of the Predictive Polygenic Risk Scores for Coronary Heart Disease Age-at-Onset in Three Different Coronary Heart Disease Family-Based Ascertainments. Circ. Genom Precis Med. 14 (3), e003201 (2021). Li, X. et al. 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 311 (Pt 1), 137039 (2023). Tian, X. et al. The association between serum Sestrin2 and the risk of coronary heart disease in patients with type 2 diabetes mellitus. BMC Cardiovasc. Disord . 22 (1), 281 (2022). Nathan, D. M. et al. Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl. J. Med. 353 (25), 2643–2653 (2005). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 12 Aug, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 27 May, 2025 Reviews received at journal 26 May, 2025 Reviewers agreed at journal 16 May, 2025 Reviews received at journal 16 May, 2025 Reviewers agreed at journal 16 May, 2025 Reviewers invited by journal 16 May, 2025 Editor assigned by journal 15 May, 2025 Editor invited by journal 30 Apr, 2025 Submission checks completed at journal 29 Apr, 2025 First submitted to journal 21 Apr, 2025 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-6492298","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":457549748,"identity":"9703f534-bb61-47b7-8f4e-303e7dbb6a26","order_by":0,"name":"Dandan Tang","email":"","orcid":"","institution":"College of Medical Engineering and Technology, Xinjiang Medical University","correspondingAuthor":false,"prefix":"","firstName":"Dandan","middleName":"","lastName":"Tang","suffix":""},{"id":457549749,"identity":"08ce9b7b-2049-4bc6-abaf-018c5dceee22","order_by":1,"name":"Fengwei Liang","email":"","orcid":"","institution":"College of Medical Engineering and Technology, Xinjiang Medical University","correspondingAuthor":false,"prefix":"","firstName":"Fengwei","middleName":"","lastName":"Liang","suffix":""},{"id":457549750,"identity":"093b034d-458f-4dfa-ad0d-2266ec26d6f0","order_by":2,"name":"Xingli Gu","email":"","orcid":"","institution":"The First Affiliated Hospital of Xinjiang Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xingli","middleName":"","lastName":"Gu","suffix":""},{"id":457549751,"identity":"73a80e6c-d703-4b61-8858-cc92d91331be","order_by":3,"name":"Yuanyuan Jin","email":"","orcid":"","institution":"College of Basic Medical Science, Xinjiang Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yuanyuan","middleName":"","lastName":"Jin","suffix":""},{"id":457549752,"identity":"92261dec-5011-47ed-8444-dac0be406457","order_by":4,"name":"Xuanjie Hu","email":"","orcid":"","institution":"College of Medical Engineering and Technology, Xinjiang Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xuanjie","middleName":"","lastName":"Hu","suffix":""},{"id":457549753,"identity":"b6e60ca6-1e45-4f43-bb8d-4907f91a3fe5","order_by":5,"name":"Fen Liu","email":"","orcid":"","institution":"Heart Center, The First Affiliated Hospital of Xinjiang Medical University","correspondingAuthor":false,"prefix":"","firstName":"Fen","middleName":"","lastName":"Liu","suffix":""},{"id":457549754,"identity":"d7ed24f7-b5c6-4519-867f-eb07578df08c","order_by":6,"name":"Yining Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIiWNgGAWjYPCCfzz87M0HDnz4QbyWAzKSPccSD87sIUGLjcENH+PDHGxEqDU4fvbwa94dd3gYbvB8OMzAwyDPL3aAgJYzeWnWvGee8TDO7t1wuMCCwXDm7AQCWg7kmBnztjHzMMuc3XB4Bg9DgsFtQlrOv4FoYZPIeXCYh40YLTdyjB/zth3m4ZHIYSBOi+SNN2aMc9vSeCR4jhkAA1mCsF/4zucYf3jbZmNvf7z58YcPP2zk+aUJaFE4wMAmxYPgS+BXDgLyDQzMH0lIJqNgFIyCUTASAQBrUkqctOIIEQAAAABJRU5ErkJggg==","orcid":"","institution":"Department of Cardiology, Xinjiang Uyghur Autonomous Region People's Hospital","correspondingAuthor":true,"prefix":"","firstName":"Yining","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-04-21 04:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6492298/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6492298/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-11142-3","type":"published","date":"2025-08-12T15:57:50+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83200601,"identity":"1379071c-31f8-4957-b864-f445af38d314","added_by":"auto","created_at":"2025-05-21 06:15:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2051155,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDescriptive plot of the original and balanced data using the SMOTENC algorithm\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/61e7f7818e70593c1a1ea77a.png"},{"id":83200602,"identity":"06d56c33-db95-4342-bc00-453c67c1c101","added_by":"auto","created_at":"2025-05-21 06:15:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2732438,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImportance of each predictor in the Lasso model. (A) Distribution of LASSO coefficients for 54 variables. (B) Selection of the optimal parameter in the LASSO regression model\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/20a363d3ceeb5cec53121fdd.png"},{"id":83199286,"identity":"4cffeec0-94b6-430c-b30e-a2d37e34ce9a","added_by":"auto","created_at":"2025-05-21 06:07:38","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":73477,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWorkflow diagram of the machine learning phase in this study\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/af92b14095f0a1f7faa07e31.png"},{"id":83199325,"identity":"f9dde1ba-d17e-4a39-bd7e-5e03bac06f31","added_by":"auto","created_at":"2025-05-21 06:07:39","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":21498593,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eROC curve analysis of five machine learning models. (A) ROC curves of five machine learning models in the unbalanced dataset; (B) ROC curves of five machine learning models in the balanced dataset\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/92d38a0912c31a5f169458fe.png"},{"id":83199291,"identity":"9260ad4f-a49d-4210-b674-40fc5c5e7793","added_by":"auto","created_at":"2025-05-21 06:07:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":6195932,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDCA plots of five machine learning models. (A) DCA plots of five machine learning models in the unbalanced dataset; (B) DCA plots of five machine learning models in the balanced dataset\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/607b4567e6058950713d6308.png"},{"id":83199293,"identity":"3a3c386e-49f9-472e-b690-3488e0f1ad24","added_by":"auto","created_at":"2025-05-21 06:07:38","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":460305,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRadar charts of five machine learning models. (A) Radar chart of five machine learning models in the unbalanced dataset; (B) Radar chart of five machine learning models in the balanced dataset\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/9004a71157bec5945c2c4c75.png"},{"id":83199311,"identity":"38f62f83-2e34-4b5a-b09c-ba7388693c93","added_by":"auto","created_at":"2025-05-21 06:07:38","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":19425196,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImportance of variables in five machine learning models trained on the balanced dataset\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/76a65cc0d983d508986fe7e5.png"},{"id":83199315,"identity":"21147f1f-f099-45cf-b9a2-1b447e1dd0ff","added_by":"auto","created_at":"2025-05-21 06:07:38","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":5960729,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSHAP explains machine learning models. (A) SHAP Summary Plot; (B) SHAP Feature Importance Plot\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-6492298/v1/b89de030714ceddc6607217c.png"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploration and Analysis of Risk Factors for Coronary Artery Disease with Type 2 Diabetes Based on SHAP Explainable Machine Learning Algorithm","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWith the accelerating aging of China's population, the phenomenon of multimorbidity in the elderly is becoming increasingly prominent. Multimorbidity refers to the simultaneous presence of two or more chronic diseases or conditions \u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. Among individuals aged 65 and above, the prevalence of multimorbidity reaches 64.7%\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e.In elderly individuals, the coexistence of coronary heart disease(Coronary Heart Disease,CHD)and diabetes (Diabetes Mellitus,DM) is one of the most common conditions. Coronary heart disease is a key condition within the spectrum of heart diseases, fundamentally caused by organic obstruction or narrowing of the coronary arteries, leading to myocardial ischemia, hypoxia, and even necrosis. Therefore, it is often referred to as ischemic heart disease \u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. Clinical manifestations include angina, arrhythmia, myocardial infarction, and even sudden death \u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. A major and modifiable risk factor for coronary heart disease that can be prevented at the population level is hyperglycemia or diabetes \u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e. In recent years, type 2 diabetes has become one of the most important complications of coronary heart disease, with its incidence gradually increasing \u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e, and it is also associated with patient mortality \u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. In terms of diagnosis, conventional diagnostic techniques include coronary angiography, coronary CT angiography (CTA), electrocardiography (ECG), and cardiac ultrasound. However, these methods require large specialized equipment and trained professionals, making them costly and inconvenient. Therefore, developing low-cost, convenient, and effective non-invasive detection methods is crucial for the early diagnosis of coronary heart disease combined with type 2 diabetes and is expected to significantly reduce patient mortality.\u003c/p\u003e \u003cp\u003eData indicate that glucose metabolism disorders are common among patients undergoing coronary angiography. Among 1,040 patients with coronary heart disease, 62.2% exhibited glucose metabolism abnormalities. The integrated management of coronary heart disease and type 2 diabetes, along with identifying patients at risk for various comorbidities, is a high priority in clinical practice \u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. Currently, machine learning algorithms have been shown to be very useful in predicting cardiovascular diseases \u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. In the medical field, the application of machine learning is infiltrating every aspect of clinical practice at an unprecedented pace, from clinical data preprocessing to patient stratification and personalized treatment plans, with its influence becoming increasingly prominent. Specifically, machine learning plays a crucial role in disease diagnosis, treatment risk assessment, drug production, and medical data analysis \u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCurrently, there is no specific model for predicting the risk of diabetes in coronary heart disease patients.This study uses machine learning algorithms to develop a clinical risk prediction model for coronary heart disease combined with type 2 diabetes.By deeply mining and integrating clinical data from patients with coronary heart disease and type 2 diabetes, and systematically analyzing the key factors contributing to the disease, this study provides solid clinical evidence for early intervention and treatment. The introduction of machine learning models will enable more accurate individual risk assessments and bring new perspectives and possibilities to the development of management strategies for this disease, showcasing broad clinical application potential.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData and Patients\u003c/h2\u003e \u003cp\u003eThis study selected detailed clinical data from a total of 29,960 cardiovascular disease patients admitted to the First Affiliated Hospital of Xinjiang Medical University between 2001 and 2018 as the research subjects. Based on the following inclusion and exclusion criteria, CHD patients and CHD-DM2 patients were selected.\u003c/p\u003e \u003cp\u003eThe inclusion criteria are as follows:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eCHD patients: Diagnosed with CHD through coronary angiography (CAG) or coronary CT angiography (CTA). Have clear clinical symptoms such as angina or other manifestations of ischemic heart disease. Aged 18 years or older.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eCHD-DM2 patients: Meet all the inclusion criteria for CHD mentioned above. Diagnosed with type 2 diabetes (based on C-peptide levels, insulin autoantibody testing, or age of onset of diabetes, among other indicators). Complete records of blood glucose control, with data on glycated hemoglobin (HbA1c) monitoring.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eExclusion criteria:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eMissing data: Patients with missing key clinical data (such as diagnostic records) or those with obvious errors.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eNon-CHD patients: Patients who have not been diagnosed as CHD. Only other types of cardiovascular diseases (such as hypertension, heart failure, arrhythmia, myocardial infarction, etc.) are present, but no CHD diagnosis.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e Patients with other serious systemic diseases: Patients with severe liver or kidney dysfunction or other systemic diseases that may affect the study results. Patients with malignant tumors who are undergoing chemotherapy or radiotherapy.\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThrough the above screening criteria, eligible CHD patients and CHD-DM2 patients were selected for further analysis and study.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMissing data imputation\u003c/h3\u003e\n\u003cp\u003eWhen the proportion of missing values in the data exceeds 20%, the data will be excluded from the final analysis dataset to ensure data integrity and analytical accuracy.。For cases where the proportion of missing values is below 20%, we will use the random forest regression method for imputation to effectively estimate and fill in the missing values, thereby maintaining the integrity of the dataset and the continuity of the analysis.\u003c/p\u003e\n\u003ch3\u003eData imbalance handling\u003c/h3\u003e\n\u003cp\u003eIn clinical data research, missing data can lead to a decline in model accuracy and even produce incorrect results. Additionally, due to the objective differences in the incidence rates of various diseases, the number of positive and negative samples is often highly imbalanced, making the class imbalance problem very common in clinical medical data, and leading to poor classification accuracy for minority class samples in the model\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. To address the data imbalance problem, this study used the SMOTENC algorithm, combined with the themis package in R (version 3.6.1) for data preprocessing. The SMOTENC algorithm is used to generate synthetic samples for the minority class to balance the class distribution in the dataset.\u003c/p\u003e\n\u003ch3\u003eFeature factor selection\u003c/h3\u003e\n\u003cp\u003eFeature selection is an important and commonly used dimensionality reduction technique, which obtains the optimal feature subset by removing irrelevant and redundant information from the dataset\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e.It is also a knowledge discovery tool that provides in-depth insights into the problem by explaining the most relevant features\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. The Lasso regression algorithm achieves dimensionality reduction and variable selection for high-dimensional data\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn this study, feature selection combined univariate analysis with Lasso regression.First, significant variables were selected through univariate analysis to initially identify potential candidate features. Then, the Lasso regression method was applied to further select these candidate features, introducing a penalty term to reduce model complexity and ultimately determine the most predictive feature set.\u003c/p\u003e\n\u003ch3\u003eModel construction\u003c/h3\u003e\n\u003cp\u003eMachine learning methods are becoming increasingly popular in medical research\u003csup\u003e[17]\u003c/sup\u003e.Supervised classification techniques are popular machine learning methods that aim to explain the dependent variable based on independent variables\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. SVM was originally introduced by Cortes and Vapnik and is now a popular classification tool in machine learning, statistics, and pattern recognition\u003csup\u003e[19\u0026ndash;20]\u003c/sup\u003e. The K-nearest neighbor algorithm (kNN) is a supervised machine learning algorithm mainly used for classification and prediction purposes\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e.Extreme Gradient Boosting (XGBoost) is a simple yet effective machine learning model, consisting of a combination of decision tree learning and gradient boosting\u003csup\u003e[22\u0026ndash;23]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe divided the included patients into a training set and a testing set in a 7:3 ratio. This study used five machine learning models (Logistic regression, Logistic_Lasso, KNN, SVM, and XGBoost) to construct prediction models. The Logistic regression, SVM, KNN, and XGBoost models were implemented using the \u0026ldquo;Caret\u0026rdquo; package in R software, while the Logistic_Lasso model was implemented using the \u0026ldquo;Caret\u0026rdquo; package combined with the \u0026ldquo;glmnet\u0026rdquo; package.The selection of model hyperparameters was performed using ten-fold cross-validation on the training dataset. Cross-validation ensures better evaluation of the model\u0026rsquo;s performance by averaging the metrics from multiple trials. The model's predictive ability was validated using the confusion matrix and the area under the receiver operating characteristic (ROC) curve (AUC).The clinical utility of the model was evaluated using the clinical decision curve analysis (DCA).\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eVariable importance assessment\u003c/h2\u003e \u003cp\u003eIn machine learning algorithms, feature importance refers to the degree to which a feature variable affects the target variable. The selection of features significantly impacts the predictive accuracy of the algorithm. An excessive or insufficient number of features can lead to overfitting or underfitting, respectively, preventing the model from achieving optimal accuracy\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003ePerform feature importance analysis on the selected model to determine the impact of each feature on the prediction. The importance of each feature is calculated based on the internal mechanisms of their respective models. The \"Caret\" package in R software is used to perform a comprehensive analysis and comparison of variable importance across different models.Additionally, we evaluated and validated the results using the optimal model.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eShapley Additive explanations (SHAP) model\u003c/h3\u003e\n\u003cp\u003eSHAP is a game-theoretic technique used to explain the output of any machine learning model\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e. In recent years, SHAP has been proposed as one of the methods for interpreting machine learning and even deep learning models. Its functions include describing the overall contribution of features, explaining their specific impact on various samples, examining feature interactions, and analyzing the joint effects of feature dependencies\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAfter selecting the best model, the SHAP model is used to display the relationships between the importance of each feature and analyze and interpret the machine learning model results. In R software, the contribution value of each variable included in the pre-designed model is displayed using histograms of varying lengths. This visualization allows for an unbiased evaluation of each variable's contribution within the system, meaning that the impact of each variable's value on the model can be independently considered and assessed\u003csup\u003e[27]\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eThe Shapiro-Wilk test is used to assess the normality of continuous variable distributions, and the Bartlett test is used to assess the homogeneity of variance in continuous variables. Continuous variables with a normal distribution are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (SD), while continuous variables with a non-normal distribution are represented by median and interquartile range (IQR). For variables that follow a normal distribution and have homogeneous variances, a t-test is used for comparison; otherwise, the Mann-Whitney U test is used for comparison. Categorical variables are expressed as percentages and analyzed using the chi-square test or Fisher's exact test. A two-sided p-value less than 0.05 is considered statistically significant. All statistical analyses are performed using R (version 3.6.1).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eBaseline analysis\u003c/h2\u003e \u003cp\u003eA total of 29,960 cardiovascular disease patients were screened according to strict inclusion and exclusion criteria, excluding 2,681 non-CHD patients and 19,657 samples with missing key clinical information. Baseline data analysis was conducted on the remaining 5,681 eligible patients. There were 4,396 patients in the CHD group, accounting for 77.3% of the sample, and 1,285 patients in the CHD-DM2 group, accounting for 22.6%. Detailed results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eIn this study, we performed a comprehensive and systematic comparative analysis of the baseline characteristics between the CHD-DM2 group and the CHD group.The results show that there are 52 indicators with differences in baseline information between the two groups.Specifically, in terms of age distribution, patients in the CHD-DM2 group were significantly younger than those in the CHD group (61 years vs. 63 years, P\u0026thinsp;\u0026lt;\u0026thinsp;0.01).The weight of patients in the CHD-DM2 group was also lower than that in the CHD group (74kg [65.0;82.0] vs. 75kg [66.0;83.0], P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The occupational distribution shows that the proportion of intellectual workers in the CHD-DM2 group is higher (26.0%), significantly greater than the 22.5% in the CHD group (P\u0026thinsp;\u0026lt;\u0026thinsp;0.01), which may be related to factors such as lifestyle, dietary habits, and work stress in specific occupational groups. In terms of hypertension prevalence, the CHD-DM2 group was significantly lower than the CHD group (48.9% vs. 62.5%, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In addition, the pulse rate in the CHD-DM2 group was lower than in the CHD group (76 times/min vs. 78 times/min, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001).Further analysis revealed that there were also significant differences between the CHD-DM2 and CHD groups in indicators such as white blood cell (WBC) count, red blood cell (RBC) count, hemoglobin (HB) content, mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).Notably, the CHD-DM2 group had lower mean platelet volume (MPV) and platelet distribution width (PDW) compared to the CHD group (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), suggesting that platelet function or activity may be regulated by the diabetic state.In addition, the CHD-DM2 group had a lower positive rate for proteinuria (Pro) and glycosuria (Glu) compared to the CHD group (4.91% vs. 8.64%; 5.14% vs. 26.3%; P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), further highlighting the specific effects of type 2 diabetes on kidney function and glucose metabolism.In terms of coagulation function, the CHD-DM2 group had a longer prothrombin time (PT) (11.3s vs. 11.1s, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while the prothrombin time activity percentage (PT.Activity) was slightly lower (102% vs. 105%, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), which may be related to an imbalance in the coagulation and fibrinolytic systems associated with diabetes.Biochemically, the CHD-DM2 group had slightly lower blood urea nitrogen (BUN) levels (5.30mmol/L vs. 5.40mmol/L, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while their uric acid (Ua) levels were higher (317\u0026micro;mol/L vs. 304\u0026micro;mol/L, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), suggesting differences in metabolic states. Finally, the blood glucose (BG) levels in the CHD-DM2 group were significantly lower than those in the CHD group (5.01mmol/L vs. 6.79mmol/L, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), directly reflecting the diagnostic features of diabetes and its stringent requirements for blood glucose management.Detailed information is provided in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of study subjects\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eALL\u003c/p\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;5681\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCHD-DM2\u003c/p\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;4396\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCHD\u003c/p\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;1285\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep.overall\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.069\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4004 (70.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3125 (71.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e879 (68.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1677 (29.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1271 (28.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e406 (31.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e61.0 [54.0;70.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61.0 [53.0;70.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e63.0 [55.0;71.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeight (cm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e170 [162;174]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170 [162;174]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e170 [162;174]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.429\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeight (kg), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e74.0 [65.0;83.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e74.0 [65.0;82.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e75.0 [66.0;83.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNational, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehan nationality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3641 (64.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2802 (63.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e839 (65.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003euyghur\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1377 (24.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1057 (24.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e320 (24.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ekazakh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e260 (4.58%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e225 (5.12%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35 (2.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehui ethnic group\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e280 (4.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e215 (4.89%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65 (5.06%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003emongolian\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60 (1.06%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44 (1.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16 (1.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eother nationalities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e63 (1.11%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e53 (1.21%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (0.78%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational.Level, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.924\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eprimary school\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1196 (21.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e928 (21.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e268 (20.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ejunior middle school\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1208 (21.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e938 (21.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e270 (21.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esenior middle school\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1654 (29.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1270 (28.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e384 (29.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003euniversity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1623 (28.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1260 (28.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e363 (28.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProfessional, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProfessional and technical personnel in institute\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1434 (25.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1145 (26.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e289 (22.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003emerchant and pasture\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1371 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1076 (24.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e295 (23.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eretirees and other personnel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2876 (50.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2175 (49.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e701 (54.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCurrent.Smoker, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.133\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enon-smoking\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3263 (57.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2501 (56.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e762 (59.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esmoking\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2418 (42.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1895 (43.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e523 (40.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCurrent.Drinker, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.149\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enon-drinking\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4460 (78.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3432 (78.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1028 (80.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003edrinking\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1221 (21.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e964 (21.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e257 (20.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertensive.history, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2729 (48.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2247 (51.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e482 (37.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2952 (52.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2149 (48.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e803 (62.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes.History. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4439 (78.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4055 (92.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e384 (29.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1242 (21.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e341 (7.76%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e901 (70.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArrhythmia, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.213\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5611 (98.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4337 (98.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1274 (99.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70 (1.23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e59 (1.34%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11 (0.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHyperlipidemia, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.749\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5635 (99.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4359 (99.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1276 (99.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e46 (0.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e37 (0.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9 (0.70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCerebral.Infarction, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.342\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5569 (98.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4314 (98.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1255 (97.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112 (1.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82 (1.87%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (2.33%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAngina, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.983\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4966 (87.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3842 (87.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1124 (87.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e715 (12.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e554 (12.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e161 (12.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeart.failure, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5664 (99.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4383 (99.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1281 (99.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (0.30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (0.30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (0.31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFamily.history, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.232\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5144 (90.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3992 (90.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1152 (89.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e537 (9.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e404 (9.19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e133 (10.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePulse (times/min), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e76.0 [69.0;83.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76.0 [69.0;82.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e78.0 [70.0;85.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWBC (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.62 [5.54;7.95]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6.54 [5.48;7.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.84 [5.72;8.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNE%, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58.4 [52.8;64.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58.3 [52.6;64.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e59.0 [53.4;64.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.062\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLY%, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31.3 [25.4;36.9]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31.4 [25.2;37.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.8 [25.5;36.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.135\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMO%, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.60 [5.40;7.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6.60 [5.40;8.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.50 [5.40;7.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.372\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEO%, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.10 [1.30;3.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.10 [1.30;3.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.10 [1.30;3.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBA%, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.30 [0.20;0.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.30 [0.20;0.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.30 [0.20;0.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.101\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNE1 (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.84 [3.05;4.85]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.78 [3.01;4.84]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.99 [3.21;4.95]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLY1 (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.01 [1.59;2.49]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.00 [1.58;2.48]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.07 [1.65;2.53]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMO1 (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.44 [0.34;0.56]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.43 [0.34;0.56]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.45 [0.35;0.58]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEO1 (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.14 [0.08;0.23]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.14 [0.08;0.23]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.14 [0.08;0.23]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBA1 (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.02 [0.01;0.04]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.02 [0.01;0.04]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.02 [0.01;0.04]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.661\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRBC (\u0026times;10\u003csup\u003e12\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.64 [4.29;4.99]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.64 [4.30;5.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.61 [4.26;4.97]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHB (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e141 [130;151]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e141 [131;151]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e139 [129;149]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHCT (L/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.48 [0.42;41.8]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.48 [0.42;41.8]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31.9 [0.42;41.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCV (fL), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e91.3 [88.5;94.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e91.5 [88.6;94.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e90.6 [87.9;93.6]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCH (pg), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30.3 [29.2;31.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.3 [29.3;31.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.1 [29.1;31.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCHC (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e331 [323;338]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e331 [323;338]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e331 [324;339]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.066\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRDW (%), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.0 [12.6;13.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13.0 [12.6;13.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13.0 [12.6;13.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.641\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePLT (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e219 [183;260]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e219 [184;261]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e217 [182;258]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.262\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMPV (fL), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10.7 [10.0;11.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10.6 [10.0;11.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.8 [10.2;11.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePCT (%), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.23 [0.20;0.28]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.23 [0.20;0.28]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.24 [0.20;0.28]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.606\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePDW (fL), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16.1 [14.6;16.9]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16.1 [14.5;16.9]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16.2 [15.0;17.1]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePro, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5354 (94.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4180 (95.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1174 (91.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003epositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e327 (5.76%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e216 (4.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e111 (8.64%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eERY, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.502\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5012 (88.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3871 (88.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1141 (88.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003epositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e669 (11.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e525 (11.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e144 (11.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlu, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5117 (90.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4170 (94.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e947 (73.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003epositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e564 (9.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e226 (5.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e338 (26.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePT (s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11.2 [10.7;11.8]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.3 [10.8;11.8]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11.1 [10.6;11.8]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePT Activity (%), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e102 [91.7;117]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e102 [91.3;115]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e105 [92.0;122]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eISR, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.02 [0.96;1.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.02 [0.96;1.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.01 [0.95;1.17]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNC, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11.2 [10.5;11.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.2 [10.5;11.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11.3 [10.6;11.5]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.219\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFIB (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.38 [2.95;3.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.36 [2.94;3.78]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.44 [3.00;3.83]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePPT (s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31.0 [29.1;33.1]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31.0 [29.2;33.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.7 [28.9;32.9]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPTT-R(s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.00 [0.90;1.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.00 [0.90;1.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00 [0.90;1.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTT Verus (s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20.2 [19.2;21.3]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20.2 [19.2;21.3]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20.2 [19.2;21.3]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.534\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPTT (s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30.8 [30.2;31.3]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.8 [30.2;31.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.7 [30.2;31.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTT (s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20.3 [20.0;20.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20.3 [20.0;20.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20.3 [20.0;20.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.632\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBUN (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.30 [4.40;6.38]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.30 [4.30;6.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.40 [4.50;6.56]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCr (\u0026micro;mol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e71.0 [61.0;82.7]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71.2 [61.0;82.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.0 [60.0;83.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUa (\u0026micro;mol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e314 [260;374]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e317 [263;377]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e304 [252;366]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBG (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.21 [4.64;6.57]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.01 [4.56;5.85]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.79 [5.38;9.29]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHbA1c (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.21 [2.01;2.48]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.16 [1.98;2.39]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.45 [2.17;2.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTG (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.50 [1.08;2.14]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.48 [1.06;2.09]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.60 [1.15;2.36]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTC (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.80 [3.10;4.56]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.83 [3.14;4.56]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.68 [2.98;4.55]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHDL (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.02 [0.85;1.23]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.03 [0.87;1.24]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.98 [0.81;1.18]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLDL (mmol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.39 [1.84;3.07]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.42 [1.87;3.09]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.28 [1.73;3.01]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPO.AI (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.13 [1.00;1.29]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.14 [1.01;1.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.12 [0.98;1.26]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPO.B (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.79 [0.62;0.98]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.79 [0.63;0.98]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.77 [0.60;0.96]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLP (mg/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e153 [85.3;278]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e154 [85.0;276]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e150 [86.6;292]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.988\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTB (\u0026micro;mol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11.0 [8.30;14.9]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.2 [8.40;15.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.6 [8.00;14.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDBIL (\u0026micro;mol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.39 [2.30;4.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.40 [2.31;4.84]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.31 [2.24;4.63]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.068\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIBIL (\u0026micro;mol/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.66 [5.10;10.8]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7.80 [5.21;11.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.23 [4.83;10.4]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTP (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e66.1 [62.7;70.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66.1 [62.6;70.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66.1 [62.8;70.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.373\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlb (g/L), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e25.8 [22.8;29.1]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25.8 [22.8;29.2]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25.9 [22.7;29.1]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.963\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAS (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.0 [30.0;35.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33.0 [30.0;35.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32.0 [30.0;35.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.128\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAVA (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20.0 [19.0;22.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20.0 [19.0;22.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20.0 [19.0;22.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLA (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34.0 [31.0;36.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33.0 [31.0;36.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34.0 [31.0;37.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLVED (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e49.0 [46.0;51.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48.0 [46.0;51.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49.0 [46.0;52.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLVES (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32.0 [30.0;34.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e32.0 [30.0;34.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32.0 [30.0;34.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.024\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIVS (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.00 [9.00;10.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.00 [9.00;10.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9.00 [9.00;10.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.675\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLVPW (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.00 [9.00;10.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.00 [9.00;10.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9.00 [9.00;10.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.274\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLVOT (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28.0 [26.0;29.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27.5 [26.0;29.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28.0 [26.0;29.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.781\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRV (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18.0 [18.0;20.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18.0 [18.0;20.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18.0 [17.0;20.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.732\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRA (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.0 [31.0;35.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33.0 [31.0;35.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33.0 [31.0;35.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.312\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePA (mm), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23.0 [21.0;24.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.0 [21.0;24.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23.0 [21.0;25.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.207\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMVE (m/s), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.71 [0.61;0.83]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.72 [0.61;0.84]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.70 [0.60;0.82]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMVA (L/min), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.73 [0.62;0.86]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.72 [0.62;0.85]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.76 [0.65;0.91]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMVEA, median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.94 [0.75;1.25]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.01 [0.76;1.26]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.85 [0.72;1.21]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFS (%), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34.0 [32.0;36.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e34.0 [33.0;36.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34.0 [32.0;36.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEF (%), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e63.0 [60.0;65.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63.0 [60.0;65.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.0 [59.0;65.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSV (%), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70.0 [63.0;78.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e70.0 [63.0;78.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.0 [63.0;78.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.606\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCO (L/min), median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.11 [4.46;5.82]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.11 [4.45;5.81]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.15 [4.51;5.86]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.108\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLTS, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5273 (92.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4103 (93.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1170 (91.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e408 (7.18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e293 (6.67%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e115 (8.95%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLAD, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2166 (38.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1793 (40.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e373 (29.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3515 (61.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2603 (59.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e912 (71.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD1.2 Narrow, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4424 (77.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3477 (79.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e947 (73.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1257 (22.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e919 (20.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e338 (26.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLCX, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3373 (59.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2749 (62.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e624 (48.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2308 (40.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1647 (37.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e661 (51.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOM, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4939 (86.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3894 (88.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1045 (81.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e742 (13.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e502 (11.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e240 (18.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRCA, no. (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3052 (53.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2480 (56.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e572 (44.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2629 (46.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1916 (43.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e713 (55.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eNote: Diabetes.History, family history of diabetes (father, mother, and/or siblings); Hypertensive.history, family history of hypertension (father, mother, and/or siblings); WBC, white blood cell; NE, neutrophil; LY, lymphocyte; MO, monocyte; EO, eosinophil; BA, basophil; NE1, neutrophil count; LY1, lymphocyte count; MO1, monocyte count; EO1, eosinophil count; BA1, basophil count; RBC, red blood cell; HB, hemoglobin; HCT, hematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; RDW, red blood cell distribution width; PLT, platelet count; MPV, mean platelet volume; PCT, platelet volume ratio; PDW, platelet distribution width; Pro, proteinuria; ERY, hematuria; Glu, glycosuria; PT, prothrombin time; PT.Activity, prothrombin time (PT) activity percentage; ISR, international standardized ratio; NC, normal control; FIB, fibrinogen; PPT, partial prothrombin time; APTT-R, activated partial thromboplastin time ratio; TT Verus, thrombin time; APTT, APTT control; TT, TT normal control; BUN, blood urea nitrogen; Cr, creatinine; Ua, uric acid; BG, blood glucose; HbA1c, glycated hemoglobin; TG, triglyceride; TC, total cholesterol; HDL, high-density lipoprotein; LDL, low-density lipoprotein; APO.AI, apolipoprotein AI; APO.B, apolipoprotein B; LP, lipoprotein; STB, total bilirubin; DBIL, direct bilirubin; IBIL, indirect bilirubin; TP, total protein; Glb, serum globulin; AS, aortic valve stenosis; AVA, valve area; LA, left atrial diameter; LVED, left ventricular end-diastolic diameter (Left Ventricular End Diastolic Diameter); LVES, left ventricular end-systolic dimension (Left Ventricular End Systolic Dimension); IVS, interventricular septal thickness; LVPW, left ventricular posterior wall thickness; LVOT, left ventricular outflow tract (Left ventricular outflow tract); RV, right ventricular diameter; RA, right atrial diameter; PA, pulmonary artery diameter; MVE, mitral valve E-wave velocity (Mitral Valve E-wave velocity); MVA, mitral valve area (Maximum Ventricular Ejection per Minute); MVEA, mitral valve E/A ratio (Mitral Valve E/A Ratio); FS, fractional shortening (Fractional Shortening); EF, left ventricular ejection fraction (Ejection Fraction); SV, stroke volume (Stroke Volume); CO, cardiac output (Cardiac Output); LTS, left main coronary artery stenosis; LAD, left anterior descending coronary artery; D1.2.Narrow, diagonal branch 1 and 2 stenosis; LCX, left circumflex coronary artery; RCA, right coronary artery\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eImbalance Handling\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows a comparison between the original dataset and the balanced dataset obtained using the SMOTENC algorithm. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the original dataset exhibits significant bias with an imbalance rate of 3.421. To address this imbalance and prevent result distortion, we used the smotenc function from the themis package in R (version 3.6.1) to balance the data. Based on statistical considerations, the value of k is set to 10, and different k values are tested for each dataset. The parameter is set to over_ratio\u0026thinsp;=\u0026thinsp;1, which represents the ratio of majority to minority frequencies. We performed oversampling on the category with the fewest observations and undersampling on the category with the most observations, thereby balancing the CHD patient group (1285) and the CHD-DM2 patient group (4396). Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e provide detailed distributions of observations for each category in the balanced and imbalanced training datasets.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescription of the original and balanced data.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDataset\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMinority class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMajority class\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSamples in total\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eImbalance rate\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCHD(original)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5681\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.421\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCHD(SMOTEN)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eFeature Selection\u003c/h2\u003e \u003cp\u003eWe adopted an integrated strategy that cleverly combines univariate analysis with Lasso regression, achieving efficient and precise feature selection. First, we used univariate analysis as a preliminary screening method, effectively narrowing down the initial set of variables to the most promising 52. During this process, we strictly controlled the error rate to ensure the robustness of the selection. Then, to further refine the feature set and enhance the model's predictive performance, we introduced Lasso regression. In the Lasso model, we implemented a strict ten-fold cross-validation strategy to comprehensively evaluate the model's performance on different data subsets. Using ten-fold cross-validation to select features with non-zero regression coefficients (λ), we successfully reduced the initially selected 52 variables to 25 key variables.The variables Age, Diabetes.History, Hypertensive.history, Weight, Pulse, LY1, MCH, PDW, Pro, Glu, PT, PT.Activity, PPT, APTT, BG, TG, APO.B, IBIL, LVES, MVA, FS, EF, LAD, LCX, and OM were identified as the final predictive factors with the greatest impact on the model's performance (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eModel Construction and Evaluation\u003c/h2\u003e \u003cp\u003eThe study population (n\u0026thinsp;=\u0026thinsp;5681) was randomly divided into a training set (n\u0026thinsp;=\u0026thinsp;3977) and a testing set (n\u0026thinsp;=\u0026thinsp;1704) at a 7:3 ratio. The training set was used to develop the predictive model, while the testing set was used for further validation. Participants were clearly classified into the CHD group (n\u0026thinsp;=\u0026thinsp;4396) and the CHD-DM2 group (n\u0026thinsp;=\u0026thinsp;1285), with the predictive model based on whether participants had CHD-DM2. The overall study workflow is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, which clearly outlines the steps of data partitioning, model construction, and validation.\u003c/p\u003e \u003cp\u003eUsing the training and testing sets, we constructed and validated five common machine learning models: Logistic, Logistic_Lasso, SVM, kNN, and XGBoost. When training the models on the balanced dataset, the accuracy, sensitivity, and specificity of all five models were significantly improved. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the XGBoost model performed particularly well on the balanced dataset. Its AUC value increased from 87.06% (95% CI: 84.85%-88.15%) on the imbalanced dataset to 95.94% (95% CI: 89.42%-91.6%), accuracy improved from 0.8656 to 0.9594, sensitivity increased from 0.7661 to 0.9054, and specificity improved from 0.8761 to 0.9150, showing superior performance compared to the other models.\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the ROC curves of five machine learning models trained on both the balanced and imbalanced datasets. All models performed better when trained on the balanced dataset. The AUC of the XGBoost model on the imbalanced dataset reached 0.8706, significantly higher than the other four models (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). On the balanced dataset, the AUC of the XGBoost model increased to 0.9594, again outperforming the other four models (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB).The DCA curve illustrates the clinical utility of the predictive model by evaluating the net benefit at different threshold probabilities \u003csup\u003e[ 28]\u003c/sup\u003e.The DCA curve (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) reveals the high clinical value of the XGBoost model, which achieves the highest net benefit at the lowest misdiagnosis rate. Overall, these findings indicate that the XGBoost model significantly outperforms the other four machine learning models, showing the most stable performance.\u003c/p\u003e \u003cp\u003eFinally, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e visually illustrates the positive impact of balancing on the dataset and model performance through a radar chart. The errors of all five machine learning models are significantly lower on the balanced dataset compared to the imbalanced dataset, further emphasizing the critical role of data balance in improving model performance.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance metrics of five machine learning models on the original and balanced datasets\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003ePrecision (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eRecall (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eImbalanced\u003c/p\u003e \u003cp\u003eData Model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.5957\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.618\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.5242\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.6441\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.2915\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.8290\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.5241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.2914\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.8500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.882\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.6828\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.9377\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.7537\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.9137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.7581\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.5912\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.8706\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.8656\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7661\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.8934\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.6674\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.9319\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.7849\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.6213\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLogistic_Lasso\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.8500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.8539\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7204\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.8761\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.6358\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.9225\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.7742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.6357\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLogistic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.8323\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.8791\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7204\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.9234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.7243\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.9220\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.7661\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.6063\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eBalanced\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003eData Model\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.6811\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.6331\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7994\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.4662\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.6003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.6985\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.5466\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.6667\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.9103\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.8324\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7426\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.9226\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.9508\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.7814\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.8213\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.8353\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.9594\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.9102\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.9054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.9150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.9144\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.9060\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.9054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.9144\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLogistic_Lasso\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.8609\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.8241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7517\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.8922\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.8749\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.7818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.7517\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.8749\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLogistic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.8523\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.8237\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.7388\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.9233\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.9045\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.7696\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.7388\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.8722\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eVariable Importance Analysis\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e illustrates the key contributions of various variables to prediction in the five predictive models trained on the balanced dataset. In the kNN model, Diabetes History, blood glucose levels, HbA1c levels, positive urine glucose test, mitral valve area, and left anterior descending artery were identified as the most important predictive factors. Similarly, for the SVM model, Diabetes History, blood glucose levels, HbA1c levels, positive urine glucose test, mitral valve area, left circumflex artery, and family history of hypertension were also considered key predictive factors. In the XGBoost model, Diabetes History, APTT ratio, International Normalized Ratio (INR), blood glucose levels, race, APTT control, and HbA1c levels are key predictive factors.On the one hand, the Logistic regression model emphasizes the high importance of Diabetes History, positive urine glucose test, Platelet Distribution Width, indirect bilirubin, obtuse margin branch, and obtuse margin branch. Furthermore, the Logistic_Lasso model emphasizes the importance of Diabetes History, positive urine glucose test, obtuse margin branch, mitral valve area level, left anterior descending coronary artery stenosis, and hypertensive family history.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn conclusion, the four variables\u0026mdash;family history of diabetes, blood glucose level, positive urine glucose test, and glycated hemoglobin\u0026mdash;are important in different models. This highlights their importance in influencing CHD-DM2 risk, indicating that these factors are key targets for risk assessment and potential intervention strategies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eModel interpretability\u003c/h2\u003e \u003cp\u003eTo more intuitively illustrate the role of the selected variables in predicting CHD-DM2, we used SHAP interpretability analysis, as shown in the summary plot (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA).In this plot, each feature importance line represents the impact of different variables on the model\u0026rsquo;s prediction. Each point corresponds to the contribution of a single patient to the prediction of that specific variable, with purple points indicating high feature values and yellow points indicating low feature values. This study identifies variables independently associated with CHD-DM2, including diabetes history, blood glucose levels, prothrombin time, glycated hemoglobin, APTT control, age, indirect bilirubin, the activity percentage of prothrombin time (PT), left atrial diameter, and platelet distribution width (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). Figure\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB shows the top ten features ranked by importance. Clearly, diabetes history, blood glucose levels, and prothrombin time are the three most critical factors driving the prediction results, while glycated hemoglobin, APTT control, age, indirect bilirubin, and the activity percentage of prothrombin time (PT) are important supplementary factors.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eOver the past decade, artificial intelligence (AI),machine learning and deep learning has become a popular subject both within and outside of the scientific community\u003csup\u003e[29]\u003c/sup\u003e.Machine learning can use sample data to build models for predicting future outcomes or identifying hidden underlying patterns in the input data \u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e. The expansion of biological data size and the increasing complexity have led to the growing application of machine learning in the field of biology \u003csup\u003e[31]\u003c/sup\u003e. Artificial intelligence (AI) methods have great potential to surpass traditional and domain-specific approaches in prediction accuracy, operational efficiency, and other aspects \u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e. Machine learning, as a powerful set of algorithms, utilizes precise identification of complex patterns in data to perform key tasks such as prediction and classification. With the increasing richness of data resources and the leap in computational power, machine learning technology has been widely applied and deeply integrated in both industry and natural sciences, demonstrating its irreplaceable value and potential \u003csup\u003e[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThis study uses the SMOTENC algorithm combined with advanced machine learning techniques to handle imbalanced data, thereby minimizing potential bias in the results\u003csup\u003e[35]\u003c/sup\u003e. By comparing the performance of five machine learning models (kNN, SVM, XGBoost, Logistic_Lasso, Logistic) before and after balancing, the results show that the data balancing strategy significantly improved the overall performance of the models, particularly the XGBoost model, which exhibited the best performance on the balanced dataset, achieving the highest precision, recall, and AUC values.\u003c/p\u003e \u003cp\u003eThe onset and progression of coronary heart disease (CHD) is a complex process influenced by both genetic and environmental factors, involving the combined effects of numerous variables. Extensive previous research has identified a range of key risk factors for CHD, including but not limited to age, gender, hypertension, diabetes, smoking habits, and dyslipidemia\u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e. To further analyze how these selected factors specifically impact the prediction model for CHD-DM2, we used SHAP to interpret the model. In explaining the model's predictions, SHAP values provide valuable information that helps us understand which features are most important in predicting coronary heart disease. This approach not only enhances the transparency of the model but also offers direction for future research and interventions\u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur study indicates that a history of diabetes, blood glucose levels, positive urine glucose test, and HbA1c are major risk factors for CHD-DM2. Given these factors, we must focus on reducing the risk of cardiovascular diseases in patients with type 2 diabetes\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e. Type 2 diabetes is a metabolic disorder characterized by chronic hyperglycemia and improper lipid, carbohydrate, and protein metabolism caused by insulin resistance and insufficient insulin secretion, leading to persistent increases in blood pressure, blood lipids, and blood glucose, thus disrupting normal body functions and metabolic balance. Epidemiological evidence indicates that any increase in blood glucose, even within the prediabetic range, increases the risk of cardiovascular diseases\u003csup\u003e[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e. In response to this situation, it is crucial to emphasize the importance of controlling blood glucose and lipid levels in CHD-DM2 patients. Regular monitoring and maintaining blood glucose within the ideal range, along with reducing the intake of high-sugar and high-fat foods, can effectively alleviate the irreversible damage caused by long-term hyperlipidemia and hyperglycemia, thereby improving patients' quality of life and delaying disease progression.\u003c/p\u003e \u003cp\u003eIn recent years, scholars and experts both domestically and internationally have generally believed that the incidence of coronary heart disease is closely related to hypertension and hyperglycemia, and based on this understanding, they have actively conducted research on treatment methods for CHD-DM2. The progress in this field urgently requires high clinical attention and should be put into practice. Specifically, when conducting clinical treatment for CHD-DM2 patients, it is crucial to monitor and record the patient's various risk factors and implement targeted interventions. With the continuous improvement of medical conditions and the deepening of related research, we believe that more scientifically effective methods to control and treat CHD-DM2 will emerge in the future, thereby providing better treatment and services for patients.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eT2DM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etype 2 diabetes\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCHD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ecoronary heart disease\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCHD-DM2\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ecoronary heart disease combined with type 2 diabetes.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics approval and consent to participate\u003c/p\u003e\n\u003cp\u003eThe study was approved by the Ethics Committee for Drug Clinical Trials of the First Affiliated Hospital of Xinjiang Medical University(Approval No. XJYKDXR20220725007). Informed consent was obtained from the participants and the study was performed in accrodance with Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003eNo conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication.I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously, and not under consideration for publication elsewhere, in whole or in part.All the authors listed have approved the manuscript that is enclosed.That its publication has been approved by the responsible authorities at the institution where the work carried out.\u003c/p\u003e\n\u003cp\u003eConsent for publication\u003c/p\u003e\n\u003cp\u003eThe co-authors agreed to publication in\u0026nbsp;“\u003cem\u003eScientific Reports\u003c/em\u003e”.The copyright to the article is transferred to\u0026nbsp;“\u003cem\u003eScientific Reports\u003c/em\u003e”\u0026nbsp;effective if and when the article is accepted for publication. The author warrants that his/her contribution is original and that he/she has full power to make this grant. The author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors.\u003c/p\u003e\n\u003cp\u003eAvailability of data and material\u003c/p\u003e\n\u003cp\u003eThis study selected clinical data from \u0026nbsp;the First Affiliated Hospital of Xinjiang Medical University. If anyone would like to obtain data from this study, please contact Dandan Tang.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eAll authors disclosed no relevant relationships.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThe present paper was supported by the following funding: (1)Youth Science Fund of the Natural Science Foundation of Xinjiang Uyghur Autonomous Region；Project Number: 2022D01C718；Project Title:\"Developing a Risk Prediction Model for Concurrent Diabetes in Coronary Heart Disease Patients Using Multimodal Data and Machine Learning\".(2)Special Funds for Talents of Xinjiang Medical University；Project Number: 0103010211.\u003c/p\u003e\n\u003cp\u003eAuthors' contributions\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDandan Tang was responsible for conceptualization, methodology, research design, data analysis, and article writing; Fengwei Liang was responsible for clinical data curation, algorithm development, SHAP implementation, statistical validation; Xingli Gu was responsible for data compilation and statistical analysis; Yuanyuan Jin was responsible for research coordination and material support; Xuanjie Hu was responsible for data collection and compilation; Fen Liu was responsible for clinical interpretation, biomarker analysis, revealing research results; Yining Yang was responsible for study design, funding acquisition, manuscript supervision.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eXu, H. et al. Establishment of a diagnostic model of coronary heart disease in elderly patients with diabetes mellitus based on machine learning algorithms. \u003cem\u003eJ. Geriatr. Cardiol.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e (6), 445\u0026ndash;455 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026auml;hler, C., Huber, C. A., Br\u0026uuml;ngger, B. \u0026amp; Reich, O. Multimorbidity, health care utilization and costs in an elderly community-dwelling population: a claims data based observational study. \u003cem\u003eBMC Health Serv. Res.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 23 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSagris, M. et al. Inflammation in Coronary Microvascular Dysfunction. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e (24), 13471 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStone, P. H., Libby, P. \u0026amp; Boden, W. E. Fundamental Pathobiology of Coronary Atherosclerosis and Clinical Implications for Chronic Ischemic Heart Disease Management-The Plaque Hypothesis: A Narrative Review. \u003cem\u003eJAMA Cardiol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e (2), 192\u0026ndash;201 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurin, T. C. et al. Diabetes and lifetime risk of coronary heart disease. \u003cem\u003ePrim. Care Diabetes\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e (5), 461\u0026ndash;466 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAronson, D. \u0026amp; Edelman, E. R. Coronary artery disease and diabetes mellitus. \u003cem\u003eCardiol. Clin.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e (3), 439\u0026ndash;455 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, Q. et al. Prediction of Atrial Fibrillation in Hospitalized Elderly Patients With Coronary Heart Disease and Type 2 Diabetes Mellitus Using Machine Learning: A Multicenter Retrospective Study. \u003cem\u003eFront. Public. Health\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e, 842104 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao, H. et al. Disease patterns of coronary heart disease and type 2 diabetes harbored distinct and shared genetic architecture. \u003cem\u003eCardiovasc. Diabetol.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (1), 276 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMirjalili, S. R. et al. An innovative model for predicting coronary heart disease using triglyceride-glucose index: a machine learning-based cohort study. \u003cem\u003eCardiovasc. Diabetol.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e (1), 200 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForrest, I. S. et al. Machine learning-based marker for coronary artery disease: derivation and validation in two longitudinal cohorts. \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e401\u003c/b\u003e (10372), 215\u0026ndash;225 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRahman, M. M. \u0026amp; Davis, D. N. Addressing the class imbalance problem in medical datasets[J]. \u003cem\u003eInt. J. Mach. Learn. Comput.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e (2), 224 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThabtah, F. et al. Data imbalance in classification: Experimental evaluation[J]. \u003cem\u003eInf. Sci.\u003c/em\u003e \u003cb\u003e513\u003c/b\u003e, 429\u0026ndash;441 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbdel Majeed, Y., Awadalla, S. S. \u0026amp; Patton, J. L. Regression techniques employing feature selection to predict clinical outcomes in stroke. \u003cem\u003ePLoS One\u003c/em\u003e. \u003cb\u003e13\u003c/b\u003e (10), e0205639 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei, H. et al. Environmental chemical exposure dynamics and machine learning-based prediction of diabetes mellitus. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cb\u003e806\u003c/b\u003e (Pt 2), 150674 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eALIZADEHSANI, R. et al. Machine learning-based coronary artery disease diagnosis: a comprehensive review [J]. \u003cem\u003eComput. Biol. Med.\u003c/em\u003e \u003cb\u003e111\u003c/b\u003e, 103346 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShao, L. et al. LASSO-derived nomogram predicting new-onset diabetes mellitus in patients with kidney disease receiving immunosuppressive drugs. J Clin Pharm Ther. ;47(10):1627\u0026ndash;1635. CHERKASSKY V.The nature of statistical learning theory [J].IEEE Trans Neural Netw, 1997, 8(6): 1564. (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlghamdi, M. et al. Predicting diabetes mellitus using SMOTE and ensemble machine learning approach: The Henry Ford ExercIse Testing (FIT) project. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e12\u003c/b\u003e (7), e0179805 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, S. et al. Applications of Support Vector Machine (SVM) Learning in Cancer Genomics. \u003cem\u003eCancer Genomics Proteom.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e (1), 41\u0026ndash;51 (2018 Jan-Feb).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, S., Sparse \u0026amp; SVM for Sufficient Data Reduction. \u003cem\u003eIEEE Trans. Pattern Anal. Mach. Intell.\u003c/em\u003e ;\u003cb\u003e44\u003c/b\u003e(9):5560\u0026ndash;5571. (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarabor, V. et al. Machine Learning Approaches for the Prediction of Hepatitis B and C Seropositivity. \u003cem\u003eInt. J. Environ. Res. Public. Health\u003c/em\u003e. \u003cb\u003e20\u003c/b\u003e (3), 2380 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, F., Yao, J., Liu, C. \u0026amp; Shou, S. Construction and validation of machine learning models for sepsis prediction in patients with acute pancreatitis. \u003cem\u003eBMC Surg.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (1), 267 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMyles, A. J., Feudaie, R. N. \u0026amp; Ln, Y. et aL An htroduction to decision tree modeling[J]. Journal of Chemometrics, 18(6): 275\u0026ndash;285 (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa, Y. et al. Explainable machine learning model reveals its decision-making process in identifying patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation. \u003cem\u003eBMC Cardiovasc. Disord\u003c/em\u003e. \u003cb\u003e23\u003c/b\u003e (1), 91 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAN Z Y, JIANG J M, CHEN, X. et al. Construction and validation of prognostic models in critically Ill patients with sepsis-associated acute kidney injury: interpretable machine learning approach[J]. \u003cem\u003eJ. translational Med.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (1), 406 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEl-Sappagh, S., Alonso, J. M., Islam, S. M. R., Sultan, A. M. \u0026amp; Kwak, K. S. A multilayer multimodal detection and prediction model based on explainable artificial intelligence for Alzheimer's disease. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (1), 2660 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMangiagalli, M., Miccolis, I., Maff\u0026eacute;, P., Pogliani, E. M. \u0026amp; Corneo, G. Role of granulocyte colony-stimulating factor in relapsed/resistant intermediate and high-grade non-Hodgkin's lymphoma patients treated with the E-SHAP regimen. Tumori. Mar-Apr;81(2):91\u0026thinsp;\u0026ndash;\u0026thinsp;5. (1995).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKondo, T. et al. Predicting Stroke in Heart Failure and Preserved Ejection Fraction Without Atrial Fibrillation. \u003cem\u003eCirc. Heart Fail.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e (7), e010377 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi, R. Y., Coyner, A. S., Kalpathy-Cramer, J., Chiang, M. F. \u0026amp; Campbell, J. P. Introduction to Machine Learning, Neural Networks, and Deep Learning. \u003cem\u003eTransl Vis. Sci. Technol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (2), 14 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThapa, R. et al. Predicting Falls in Long-term Care Facilities: Machine Learning Study. \u003cem\u003eJMIR Aging\u003c/em\u003e. \u003cb\u003e5\u003c/b\u003e (2), e35373 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreener, J. G., Kandathil, S. M., Moffat, L. \u0026amp; Jones, D. T. A guide to machine learning for biologists. \u003cem\u003eNat. Rev. Mol. Cell. Biol.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (1), 40\u0026ndash;55 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartin-Morales, A. et al. Predicting Cardiovascular Disease Mortality: Leveraging Machine Learning for Comprehensive Assessment of Health and Nutrition Variables. \u003cem\u003eNutrients\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e (18), 3937 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, H. et al. PTX3 mediates the infiltration, migration, and inflammation-resolving-polarization of macrophages in glioblastoma. \u003cem\u003eCNS Neurosci. Ther.\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e (11), 1748\u0026ndash;1766 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu, M. et al. A Risk Prediction Model Based on Machine Learning for Cognitive Impairment Among Chinese Community-Dwelling Elderly People With Normal Cognition: Development and Validation Study. \u003cem\u003eJ. Med. Internet Res.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e (2), e20298 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThakur, V. S., Kankar, P. K., Parey, A., Jain, A. \u0026amp; Jain, P. K. The implication of oversampling on the effectiveness of force signals in the fault detection of endodontic instruments during RCT. \u003cem\u003eProc. Inst. Mech. Eng. H\u003c/em\u003e. \u003cb\u003e237\u003c/b\u003e (8), 958\u0026ndash;974 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeitosa, M. F. et al. Heterogeneity of the Predictive Polygenic Risk Scores for Coronary Heart Disease Age-at-Onset in Three Different Coronary Heart Disease Family-Based Ascertainments. \u003cem\u003eCirc. Genom Precis Med.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (3), e003201 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, X. et al. 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. \u003cem\u003eChemosphere\u003c/em\u003e \u003cb\u003e311\u003c/b\u003e (Pt 1), 137039 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTian, X. et al. The association between serum Sestrin2 and the risk of coronary heart disease in patients with type 2 diabetes mellitus. \u003cem\u003eBMC Cardiovasc. Disord\u003c/em\u003e. \u003cb\u003e22\u003c/b\u003e (1), 281 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNathan, D. M. et al. Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. \u003cem\u003eN Engl. J. Med.\u003c/em\u003e \u003cb\u003e353\u003c/b\u003e (25), 2643\u0026ndash;2653 (2005).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":false,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Coronary heart disease combined with type 2 diabetes, Machine learning, SHAP, Imbalance processing","lastPublishedDoi":"10.21203/rs.3.rs-6492298/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6492298/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e Type 2 diabetes (T2DM) significantly elevates coronary heart disease (CHD) risk. This study leverages interpretable machine learning (ML) to identify risk factors for CHD with T2DM, enhancing clinical decision-making.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eClinical data from 5,681 cardiovascular patients (4,396 CHD; 1,285 CHD+T2DM) hospitalized between 2001-2018 were analyzed. The SMOTENC algorithm addressed dataset imbalance. Predictive variables were selected via univariate analysis and Lasso regression. Five ML models (logistic regression, Lasso regression, KNN, SVM, XGBoost) were developed and validated using accuracy, sensitivity, specificity, ROC, and decision curve analysis. SHAP values interpreted model outputs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eData were split into training (n=3,977) and validation (n=1,704) sets. Lasso regression identified 25 predictive variables. XGBoost achieved superior performance (highest accuracy: 0.89; AUC: 0.93) and net benefit in decision curves. SHAP analysis revealed diabetes duration, blood glucose (BG), prothrombin time (PT), and glycated hemoglobin (HbA1c) as primary risk factors. Positive urine glucose and elevated low-density lipoprotein also contributed significantly.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eDiabetes history, BG, HbA1c, and PT are critical risk factors for CHD-T2DM comorbidity. Prioritizing monitoring of these parameters and implementing targeted interventions may mitigate risk. The XGBoost-SHAP framework provides an interpretable tool for clinical risk stratification.\u003c/p\u003e","manuscriptTitle":"Exploration and Analysis of Risk Factors for Coronary Artery Disease with Type 2 Diabetes Based on SHAP Explainable Machine Learning Algorithm","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-21 06:07:33","doi":"10.21203/rs.3.rs-6492298/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-05-27T04:22:16+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-05-26T10:47:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"198482789815089383186063516015166236646","date":"2025-05-16T09:56:06+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-05-16T08:49:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"298246404414036809261381601259630143125","date":"2025-05-16T08:21:14+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-05-16T07:32:37+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-15T14:30:22+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-04-30T12:24:24+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-29T17:36:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-04-21T04:19:31+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3b02a950-bb17-46e5-9556-ebd9feb4c8cf","owner":[],"postedDate":"May 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":48627734,"name":"Biological sciences/Biological techniques/Biological models"},{"id":48627735,"name":"Health sciences/Cardiology"},{"id":48627736,"name":"Health sciences/Diseases"}],"tags":[],"updatedAt":"2025-08-18T16:06:05+00:00","versionOfRecord":{"articleIdentity":"rs-6492298","link":"https://doi.org/10.1038/s41598-025-11142-3","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-08-12 15:57:50","publishedOnDateReadable":"August 12th, 2025"},"versionCreatedAt":"2025-05-21 06:07:33","video":"","vorDoi":"10.1038/s41598-025-11142-3","vorDoiUrl":"https://doi.org/10.1038/s41598-025-11142-3","workflowStages":[]},"version":"v1","identity":"rs-6492298","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6492298","identity":"rs-6492298","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}