研究目的
To identify the best predictors of HDL and LDL cholesterol using statistical analyses and two machine learning algorithms and to compare the predictive power of combined anthropometric measures in Korean adults.
研究成果
The best predictors of the HDL cholesterol were the rib to hip ratio in women and the WHR in men, whereas the strongest indicators of the LDL cholesterol among all of the variables were age in women and BMI in men. A combination of anthropometric measures could slightly improve the prediction of both HDL and LDL cholesterol levels. The findings provide information for the development of better initial screening tools for HDL and LDL cholesterol levels.
研究不足
The study could not establish a cause-and-effect relationship due to the nature of the cross-sectional study design. The results were limited by country and by ethnic group because of the participation of only Korean men and women. The sensitivity of the methods was low in several prediction experiments because of the lack of subjects with hypo-HDL and hyper-LDL.
1：Experimental Design and Method Selection:
The study used binary logistic regression (LR) and the naive Bayes algorithm (NB) to evaluate statistically significant differences between subjects with normal and high LDL cholesterol levels and between subjects with normal and low HDL cholesterol levels.
2：Sample Selection and Data Sources:
A total of 13,014 subjects (7,662 women and 5,352 men) aged 20–90 years old participated in this study. Data were obtained from the Korean Health and Genome Epidemiology study database.
3：List of Experimental Equipment and Materials:
Anthropometric characteristics were measured by well-trained technicians based on standardized protocols. Body weight and height were measured with the subject in lightweight clothing and without shoes to the nearest 0.1 kg and 0.1 cm, respectively (LG-150; G Tech International Co., Ltd., Uijeongbu). The body circumferences were measured at eight sites (forehead, neck, axilla, chest, ribs, waist, pelvis, and hips) using nonelastic tape.
4：1 kg and 1 cm, respectively (LG-150; G Tech International Co., Ltd., Uijeongbu). The body circumferences were measured at eight sites (forehead, neck, axilla, chest, ribs, waist, pelvis, and hips) using nonelastic tape. Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: The anthropometric measures were assessed with binary logistic regression (LR) to evaluate statistically significant differences. LR and the naive Bayes algorithm (NB) were used in the analyses of the predictive power of individual and combined measures.
5：Data Analysis Methods:
All the statistical analyses and prediction experiments were performed in SPSS 19 for Windows (SPSS, Inc., Chicago, IL, USA) and Weka (the Waikato Environment for Knowledge Analysis data mining tool).
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
