Abstract
ABSTRACTBackgroundAs the elderly population gradually increases, musculoskeletal disorders such as sarcopenia are increasing. Diagnostic techniques such as X-ray, CT, and MRI imaging are used to predict and diagnose sarcopenia, and methods using machine learning are gradually increasing.PurposeThe purpose of this study was to create a model that can predict sarcopenia using physical characteristics and activity-related variables without medical diagnostic equipment such as imaging equipment for the elderly aged 60 years or older.MethodA sarcopenia prediction model was constructed using public data obtained from the Korea National Health and Nutrition Examination Survey. Models were built using the multi-layer perceptron, XGBoost, LightGBM, and RandomForest algorithms, and the feature importance of the model with the highest accuracy was analyzed through evaluation metrics.ResultThe sarcopenia prediction model built with the LightGBM algorithm showed the highest test accuracy at 0.852. In constructing the LightGBM model, physical characteristics variables such as BMI showed high importance, and activity-related variables were also used in constructing the model.ConclusionThe sarcopenia prediction model composed only of physical characteristics and activity-related factors showed excellent performance, and the use of this model will help predict sarcopenia in the elderly living in communities with insufficient medical resources or difficult access to medical facilities.
Publisher
Cold Spring Harbor Laboratory
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献