Postgraduate Program in Nursing, Federal University of Pelotas. Pelotas, Rio Grande do Sul, Brazil; Postgraduate Program in Public Health Nursing, University of São Paulo, Ribeirão Preto, Brazil.
Postgraduate Program in Nursing, Federal University of Pelotas. Pelotas, Rio Grande do Sul, Brazil.
Nutr Metab Cardiovasc Dis. 2024 Sep;34(9):2034-2045. doi: 10.1016/j.numecd.2024.05.020. Epub 2024 May 29.
Machine learning may be a tool with the potential for obesity prediction. This study aims to review the literature on the performance of machine learning models in predicting obesity and to quantify the pooled results through a meta-analysis.
A systematic review and meta-analysis were conducted, including studies that used machine learning to predict obesity. Searches were conducted in October 2023 across databases including LILACS, Web of Science, Scopus, Embase, and CINAHL. We included studies that utilized classification models and reported results in the Area Under the ROC Curve (AUC) (PROSPERO registration: CRD42022306940), without imposing restrictions on the year of publication. The risk of bias was assessed using an adapted version of the Transparent Reporting of a multivariable prediction model for individual Prognosis or Diagnosis (TRIPOD). Meta-analysis was conducted using MedCalc software. A total of 14 studies were included, with the majority demonstrating satisfactory performance for obesity prediction, with AUCs exceeding 0.70. The random forest algorithm emerged as the top performer in obesity prediction, achieving an AUC of 0.86 (95%CI: 0.76-0.96; I: 99.8%), closely followed by logistic regression with an AUC of 0.85 (95%CI: 0.75-0.95; I: 99.6%). The least effective model was gradient boosting, with an AUC of 0.77 (95%CI: 0.71-0.82; I: 98.1%).
Machine learning models demonstrated satisfactory predictive performance for obesity. However, future research should utilize more comparable data, larger databases, and a broader range of machine learning models.
机器学习可能是一种具有肥胖预测潜力的工具。本研究旨在综述机器学习模型在肥胖预测中的应用文献，并通过荟萃分析量化汇总结果。
进行了系统评价和荟萃分析，纳入了使用机器学习预测肥胖的研究。2023 年 10 月，在 LILACS、Web of Science、Scopus、Embase 和 CINAHL 等数据库中进行了检索。我们纳入了使用分类模型且报告了受试者工作特征曲线下面积（AUC）（PROSPERO 注册：CRD42022306940）结果的研究，不限制发表年份。使用改良版的多变量个体预后或诊断预测模型透明报告（TRIPOD）评估偏倚风险。使用 MedCalc 软件进行荟萃分析。共纳入 14 项研究，其中大多数研究显示肥胖预测性能良好，AUC 超过 0.70。随机森林算法在肥胖预测中表现最佳，AUC 为 0.86（95%CI：0.76-0.96；I²：99.8%），其次是逻辑回归，AUC 为 0.85（95%CI：0.75-0.95；I²：99.6%）。表现最差的模型是梯度提升，AUC 为 0.77（95%CI：0.71-0.82；I²：98.1%）。
机器学习模型在肥胖预测中表现出良好的预测性能。然而，未来的研究应利用更具可比性的数据、更大的数据库和更广泛的机器学习模型。
