Background: Coronary artery calcium score (CACS) is a reliable predictor for future cardiovascular disease risk. Although deep learning studies using computed tomography (CT) images to predict CACS have been reported, no study has assessed the feasibility of machine learning (ML) algorithms to predict the CACS using clinical variables in a healthy general population. Therefore, we aimed to assess whether ML algorithms other than binary logistic regression (BLR) could predict high CACS in a healthy population with general health examination data. Methods: This retrospective observational study included participants who had regular health screening including coronary CT angiography. High CACS was defined by the Agatston score ≥ 100. Univariable and multivariable BLR was performed to assess predictors for high CACS in the entire dataset. When performing ML prediction for high CACS, the dataset was randomly divided into a training and test dataset with a 7:3 ratio. BLR, catboost, and xgboost algorithms with 5-fold cross-validation and grid search technique were used to find the best performing classifier. Performance comparison of each ML algorithm was evaluated with the area under the receiver operating characteristic (AUROC) curve. Results: A total of 2133 participants were included in the final analysis. Mean age and proportion of male sex were 55.4 ± 11.3 years and 1483 (69.5%), respectively. In multivariable BLR analysis, age (odds ratio [OR], 1.12; 95% confidence interval [CI], 1.10–1.15, p < 0.001), male sex (OR, 2.91; 95% CI, 1.57–5.38, p < 0.001), systolic blood pressure (OR, 1.02; 95% CI, 1.00–1.03, p = 0.019), and low-density lipoprotein cholesterol (OR, 1.00; 95% CI, 0.99–1.00, p = 0.047) were significant predictors for high CACS. Performance in predicting high CACS of xgboost was AUROC of 0.823, followed by catboost (0.750) and BLR (0.585). The comparison of AUROC between xgboost and BLR was significant (p for AUROC comparison < 0.001). Conclusions: Xgboost ML algorithm was found to be a more reliable predictor of CACS in healthy participants compared to the BLR algorithm. ML algorithms may be useful for predicting CACS with only laboratory data in healthy participants.

As the association between obesity and bone health remains controversial in children and adolescents, we investigate the effects of obesity parameters on bone mineral density (BMD) in 2060 Korean adolescents who participated in the 2008–2011 Korea National Health and Nutrition Examination Survey (KNHANES). Multiple regression analysis and analysis of covariance (ANCOVA) were conducted to examine both the linear and non-linear associations between total-body-less-head (TBLH) BMD and four obesity parameters: body mass index, waist circumference, waist-to-height ratio, and total-body fat mass (FM). In a multiple linear regression analysis adjusted for age, menarcheal status (in females only), and total-body lean mass, there was no significant linear association between obesity parameters and TBLH BMD, except for total-body FM in males. However, upon adding a second-order polynomial term for each obesity parameter, a significant quadratic relationship between all obesity parameters and TBLH BMD was observed, with the corresponding quadratic term being negative. The results of ANCOVA also revealed an inverted U-shaped relationship between each obesity parameter and TBLH BMD. Our findings suggest the existence of an optimal range of obesity parameters for developing or maintaining optimal bone health in Korean Adolescents. Deviation from this range, in either direction (being underweight or having obesity), may compromise bone health.

Vascular calcification is a major contributor to mortality in end-stage renal disease (ESRD) patients. In this study, we investigated whether there was a correlation between the coronary artery calcium score (CACS) and the vascular calcification score (VCS), and whether higher VCS increased the incidence of interventions and major adverse cardiac and cerebrovascular events (MACCE). ECG-gated CT, including vascular access and the coronary vessel, was taken. CACS and VCS were calculated by the Agatston method. A comparison of CACS and survival analysis according to VCS groups was performed. Using a cutoff of VCS = 500, 77 patients were divided into two groups. The vintage was significantly older in the higher VCS group. The median CACS was higher in the higher VCS group (21 [0, 171] vs. 552 [93, 2430], p < 0.001). The hazard ratio (HR) for interventions and MACCEs in the higher VCS group increased by 3.2 and 2.3, respectively. Additionally, a longer duration of hemodialysis and higher magnesium levels (>2.5 mg/dL) showed lower HRs for interventions (<1). We quantified VCS and found that it was associated with the CACS. Additionally, higher VCS increased the risk of access interventions and MACCE. VCS of the access site may be suggested as a biomarker to predict ESRD patients.