ORIGINAL RESEARCH article
Front. Public Health
Sec. Aging and Public Health
Volume 13 - 2025 | doi: 10.3389/fpubh.2025.1576338
This article is part of the Research TopicExploring the Impact of Nutrition and Physical Activity on Sarcopenic ObesityView all 6 articles
Prediction Models for Sarcopenic Obesity Based on Machine Learning
Provisionally accepted
- The Affiliated Hospital of Qingdao University, Qingdao, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Background: Facing the accelerating progress of global aging, sarcopenic obesity (SO) in older adults, is associated with significant rates of disability and mortality.SO has become a serious and important public health problem. This study aims to develop and validate predictive models using machine learning (ML) to identify SO in patients. Method: The data of 386 participants collected from the Affiliated Hospital of Qingdao University are divided in an 8:2 ratio, with 80% used for training and 20% for test.We use univariate analysis to identify the correlation factors of SO, and then use multivariate logistic regression analysis to determine the independent influencing factors of SO. The SHAP diagram is used to clarify the importance of variables in the model. the Predictive models for SO, we used five models and internal five-fold cross-verification to determine the most suitable hyperparameter for the model. Result: In 386 participants, 61 are diagnosed with sarcopenic obesity(15.8%). We select four independent predictive factors: BMI, bather index score, grip strength, and calf circumference that calf circumference plays an important role in the risk of SO in older adult. the AUC values of the test set for Random Forest(RF), Naive Bayes(NB), Light Gradient Boosting Machine (LightGBM), K-nearest Neighbor Algorithm(KNN), and Extreme Gradient Boosting(XGBoost) are recorded as 0.839, 0.815, 0.808, 0.794, and 0.798. The average performance of the machine learning model in the training set is the best for the RF model (AUC: 0.839). Conclusion: We construct a predictive model based on the results of the RF model that combine four clinical predictors (BMI, bather index score, grip strength, and calf circumference) to reliably predict SO.
Keywords: Sarcopenia, RF, Calf circumference, Obesity, machine learning
Received: 13 Feb 2025; Accepted: 27 May 2025.
Copyright: © 2025 徐, LIU, Hu, Luan, Duan, Wang, 崔, Zhou and Mao. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Yongjun Mao, The Affiliated Hospital of Qingdao University, Qingdao, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.