Dataset Information

Development of a Metabolic Syndrome Classification and Prediction Model for Koreans Using Deep Learning Technology: The Korea National Health and Nutrition Examination Survey (KNHANES) (2013-2018).

ABSTRACT: The prevalence of metabolic syndrome (MetS) and its cost are increasing due to lifestyle changes and aging. This study aimed to develop a deep neural network model for prediction and classification of MetS according to nutrient intake and other MetS-related factors. This study included 17,848 individuals aged 40-69 years from the Korea National Health and Nutrition Examination Survey (2013-2018). We set MetS (3-5 risk factors present) as the dependent variable and 52 MetS-related factors and nutrient intake variables as independent variables in a regression analysis. The analysis compared and analyzed model accuracy, precision and recall by conventional logistic regression, machine learning-based logistic regression and deep learning. The accuracy of train data was 81.2089, and the accuracy of test data was 81.1485 in a MetS classification and prediction model developed in this study. These accuracies were higher than those obtained by conventional logistic regression or machine learning-based logistic regression. Precision, recall, and F1-score also showed the high accuracy in the deep learning model. Blood alanine aminotransferase (β = 12.2035) level showed the highest regression coefficient followed by blood aspartate aminotransferase (β = 11.771) level, waist circumference (β = 10.8555), body mass index (β = 10.3842), and blood glycated hemoglobin (β = 10.1802) level. Fats (cholesterol [β = -2.0545] and saturated fatty acid [β = -2.0483]) showed high regression coefficients among nutrient intakes. The deep learning model for classification and prediction on MetS showed a higher accuracy than conventional logistic regression or machine learning-based logistic regression.

SUBMITTER: Kim H

PROVIDER: S-EPMC10193438 | biostudies-literature | 2023 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development of a Metabolic Syndrome Classification and Prediction Model for Koreans Using Deep Learning Technology: The Korea National Health and Nutrition Examination Survey (KNHANES) (2013-2018).

Kim Hyerim H Heo Ji Hye JH Lim Dong Hoon DH Kim Yoona Y

Clinical nutrition research 20230425 2

The prevalence of metabolic syndrome (MetS) and its cost are increasing due to lifestyle changes and aging. This study aimed to develop a deep neural network model for prediction and classification of MetS according to nutrient intake and other MetS-related factors. This study included 17,848 individuals aged 40-69 years from the Korea National Health and Nutrition Examination Survey (2013-2018). We set MetS (3-5 risk factors present) as the dependent variable and 52 MetS-related factors and nut ...[more]

PMID: 37214780

Similar Datasets

Project description:BackgroundArsenic is known as an endocrine disruptor that people are exposed to through various sources such as drinking water and indigestion of marine products. Although some epidemiological and animal studies have reported a correlation between arsenic exposure and diabetes development, there are limited studies regarding the toxic effects of organic arsenic including arsenobetaine on the human body. Here, we analyzed the association between urine arsenobetaine and the homeostasis model assessment of β-cell function (HOMA-β), which is an index for predicting diabetes development and reflecting the function of pancreatic β-cells.MethodsIn the fourth Korea National Health and Nutrition Examination Survey (KNHANES), health and nutrition surveys and screening tests were performed. Of the total survey population, people with confirmed values for urine total arsenic and arsenobetaine were included, and known diabetic patients were excluded. A total 369 participants were finally included in the study. We collected surveys on health, height, body weight, body mass index, blood mercury level, fasting glucose level, and serum insulin level and calculated HOMA index. Owing to sexual discrepancy, we performed sexually stratified analysis.ResultsUrine total arsenic and total arsenic minus arsenobetaine was not associated with HOMA-IR and HOMA-β in univariate analysis or in sexually stratified analysis. However, urine arsenobetaine showed a statistically significant relationship with HOMA-β in univariate analysis, and only male participants showed a significant correlation in sexually stratified analysis. In the analysis adjusted for age, BMI, smoking, alcohol drinking, physical activity and blood mercury, the HOMA-β value in the group below the 25th percentile of arsenobetaine was significantly higher than the group between 50 and 75th percentile, while no difference was shown for HOMA-IR. In sexually stratified analysis, The value of HOMA-β was significantly higher in male participants with below the 25th percentile urine arsenobetaine than the group between 25 and 50th and between 50 and 75th, while no difference was shown for HOMA-IR. However, female participants did not demonstrate a relationship between HOMA-IR, HOMA-β and urine arsenobetaine.ConclusionThis study revealed the association between urine arsenobetaine and pancreatic β-cell function assessed by HOMA-β in the normal population (without diabetes), especially in males, despite adjusting for factors affecting pancreatic β-cell function and diabetes.

Dataset Information

Development of a Metabolic Syndrome Classification and Prediction Model for Koreans Using Deep Learning Technology: The Korea National Health and Nutrition Examination Survey (KNHANES) (2013-2018).

Publications

Development of a Metabolic Syndrome Classification and Prediction Model for Koreans Using Deep Learning Technology: The Korea National Health and Nutrition Examination Survey (KNHANES) (2013-2018).

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets