Department of International Healthcare Administration, Daegu Catholic University, Gyeongsan 38430, Korea.
College of Pharmacy, Keimyung University, Daegu 42601, Korea.
Int J Environ Res Public Health. 2019 Jul 25;16(15):2666. doi: 10.3390/ijerph16152666.
Lead, mercury, and cadmium are common environmental pollutants in industrialized countries, but their combined impact on hypercholesterolemia (HC) is poorly understood. The aim of this study was to compare the performance of various machine learning (ML) models to predict the prevalence of HC associated with exposure to lead, mercury, and cadmium. A total of 10,089 participants of the Korea National Health and Nutrition Examination Surveys 2008-2013 were selected and their demographic characteristics, blood concentration of metals, and total cholesterol levels were collected for analysis. For prediction, five ML models, including logistic regression (LR), k-nearest neighbors, decision trees, random forests, and support vector machines (SVM) were constructed and their predictive performances were compared. Of the five ML models, the SVM model was the most accurate and the LR model had the highest area under receiver operating characteristic (ROC) curve of 0.718 (95% CI: 0.688-0.748). This study shows the potential of various ML methods to predict HC associated with exposure to metals using population-based survey data.
铅、汞和镉是工业化国家常见的环境污染物，但它们对高胆固醇血症（HC）的综合影响尚不清楚。本研究旨在比较各种机器学习（ML）模型在预测与铅、汞和镉暴露相关的 HC 患病率方面的性能。从 2008-2013 年韩国国家健康和营养检查调查中选择了 10089 名参与者，并收集了他们的人口统计学特征、金属血液浓度和总胆固醇水平进行分析。在预测方面，构建了包括逻辑回归（LR）、k-最近邻、决策树、随机森林和支持向量机（SVM）在内的五个 ML 模型，并比较了它们的预测性能。在这五个 ML 模型中，SVM 模型是最准确的，LR 模型的接收者操作特征（ROC）曲线下面积最高，为 0.718（95%置信区间：0.688-0.748）。这项研究表明，使用基于人群的调查数据，各种 ML 方法在预测与金属暴露相关的 HC 方面具有潜力。
