通过使用机器学习方法通过非饮食生活方式因素预测早期肥胖风险
Ker Ming Seaw1, Melvin Khee Shing Leow1,2,3,4,5, Xinyan Bi1
1Clinical Nutrition Research Centre (CNRC), Singapore Institute of Food and Biotechnology Innovation (SIFBI), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore.
Clinical obesity
|April 3, 2025
概括
预测肥胖风险是可以使用非饮食因素,如生活方式和家族史. 机器学习模型,特别是随机森林,显示出早期识别和预防肥胖和相关非传染性疾病 (NCD) 的潜力.
科学领域:
- 公共卫生 公共卫生
- 生物医学信息学 生物医学信息学
- 医疗保健中的机器学习
背景情况:
- 肥胖是非传染性疾病 (NCD) 的一个主要风险因素.
- 早期发现肥胖风险对于有效的预防策略至关重要.
- 非饮食因素需要进一步调查作为肥胖预测因素.
研究的目的:
- 评估非饮食因素 (生活方式,家族史,人口统计) 对肥胖风险的预测能力.
- 为了比较各种机器学习模型在预测肥胖方面的表现.
- 探索肥胖预测模型的潜力,以尽量减少对饮食数据的依赖.
主要方法:
- 利用了211个人的数据集 (1068名男性,1043名女性,年龄在14-61岁之间).
- 开发和评估机器学习模型,包括决策树,随机森林,支持矢量分类 (SVC),K-最近邻居 (KNN) 和高斯素朴湾 (GNB).
- 专注于用于模型训练和预测的非饮食因素,使用准确性,精度,回忆,F1得分,特异性和AUC-ROC等指标评估性能,并启动可变性分析.
主要成果:
- 随机森林被确定为最佳模型,达到66.9%的测试准确度,66.4%的精度,66.9%的回忆,66.4%的F1得分,94.5%的特异性和92.3%的AUC-ROC.
- 非饮食因素,包括家族史和人口统计数据,显示出肥胖风险的显著预测能力.
- 生活方式因素也有助于预测准确度,尽管比家族史和人口统计数据要少.
结论:
- 非饮食因素是肥胖风险的可行预测因素,为传统的饮食评估提供了替代方案.
- 机器学习模型,特别是随机森林,可以有效地识别患肥胖风险的个体.
- 这种方法通过突出关键的生活方式和人口预测因素,支持早期干预和预防肥胖及其相关的非传染性疾病.
相关概念视频
Obesity
348
The Body Mass Index (BMI) is a numerical value derived from a person's weight and height, used to categorize individuals into weight ranges. It is calculated using the formula: weight in kilograms divided by height in meters squared. Obesity is a health condition characterized by excessive accumulation of adipose tissue that poses health risks, often diagnosed with a BMI ≥ 30. This excess fat storage occurs when surplus dietary calories are converted into triglycerides and stored in...
348
Cancer Prevention
6.0K
Several factors can increase the risk of cancer in an individual. About 50% of cancer cases can be prevented by adopting a healthy lifestyle, regular exercise, eating healthy, and following a modest cancer prevention diet. Epidemiological studies have consistently shown that populations with vegetable and fruit-rich diets have reduced the incidence of cancer. On the other hand, populations who have a diet rich in animal fat, red meat, junk food, or high calories are predisposed to cancer.
Some...
Some...
6.0K
Nature and Nurture
20.3K
Many human characteristics, like height, are shaped by both nature—in other words, by our genes—and by nurture, or our environment. For example, chronic stress during childhood inhibits the production of growth hormones and consequently reduces bone growth and height. Scientists estimate that 70-90% of variation in height is due to genetic differences among individuals, and 10-30% of variation in height is due to differences in the environments that individuals experience,...
20.3K


