Construction of a risk prediction model for sarcopenic obesity in the elderly based on machine learning

Fan Qingqing; Peng Peng; Li Yaling; Yang Peiji; Zhou Yilin; Tang Haomin; Mao Shuangyan; Deng Dan

doi:10.13406/j.cnki.cyxb.004107

您当前的位置：

首页 >

文章列表页 >

Construction of a risk prediction model for sarcopenic obesity in the elderly based on machine learning

Clinical research | 更新时间：2026-06-11

- Construction of a risk prediction model for sarcopenic obesity in the elderly based on machine learning
- Journal of Chongqing Medical University Vol. 51, Issue 6, Pages: 867-875(2026)
- 作者机构：
  
  1.重庆医科大学公共卫生学院卫生统计教研室，重庆 400016
  2.重庆市结核病防治所，重庆 400050
  3.重庆医科大学附属第一医院老年医学科，重庆 400016
- 作者简介：
- 基金信息：
- DOI：10.13406/j.cnki.cyxb.004107
  CLC： R589.2
- Received：21 January 2026，
  
  Revised：2026-04-07，
  
  Accepted：08 April 2026，
  
  Published：28 June 2026
- 稿件说明：
移动端阅览
范清清，彭朋，李亚玲，等.基于机器学习构建老年肌少性肥胖风险预测模型[J].重庆医科大学学报，2026，51（6）：867-875.

Fan Qingqing,Peng Peng,Li Yaling,et al.Construction of a risk prediction model for sarcopenic obesity in the elderly based on machine learning[J].Journal of Chongqing Medical University,2026,51(6):867-875.
范清清，彭朋，李亚玲，等.基于机器学习构建老年肌少性肥胖风险预测模型[J].重庆医科大学学报，2026，51（6）：867-875. DOI： 10.13406/j.cnki.cyxb.004107.

Fan Qingqing,Peng Peng,Li Yaling,et al.Construction of a risk prediction model for sarcopenic obesity in the elderly based on machine learning[J].Journal of Chongqing Medical University,2026,51(6):867-875. DOI： 10.13406/j.cnki.cyxb.004107.

摘要

目的

基于机器学习探究老年住院患者肌少性肥胖的影响因素，构建较优的风险预测模型，为早期筛查肌少性肥胖提供工具。

方法

回顾性收集2016年3月至2023年6月于重庆医科大学附属第一医院就诊的老年住院患者数据，基于单因素分析和最小绝对收缩和选择算子（least absolute shrinkage and selection operator，LASSO）回归筛选变量，多因素logistic回归探究肌少性肥胖影响因素，构建logistic回归、极端梯度提升（eXtreme gradient boosting，XGBoost）、随机森林（random forest，RF）和反向传播神经网络（back propagation neural network，BPNN）4种预测模型，识别最优模型，并采用沙普利值可加性解释（shapley additive explanation，SHAP）进行模型解释。

结果

老年住院患者肌少性肥胖检出率为26.41%。衰弱前期［比值比（odds ratio，

）=2.493］、衰弱（

=4.091）、经年龄校正的共病指数（age-adjusted Charlson comorbidity index，aCCI）（

=1.110）、认知障碍（

=1.482）、体脂百分比（

=1.114）和内脏脂肪面积（

=1.014）是老年住院患者肌少性肥胖的独立危险因素（

0.05）。BPNN的预测效能最优，模型测试集受试者工作特征曲线下面积（area under the curve，AUC）为0.842，Brier分数0.144，模型性能良好，临床决策曲线显示在阈值范围内，模型获益最大。SHAP可视化显示特征重要性前三依次是内脏脂肪面积、无衰弱和体脂百分比。

结论

构建BPNN预测模型，为识别肌少性肥胖高危人群提供方法学参考。建议重点关注患有认知障碍、体脂百分比≥35%、内脏脂肪面积≥100 cm

、衰弱评估异常的老年住院患者，并对高危人群开展综合性干预。

Abstract

Objective

To investigate the influencing factors for sarcopenic obesity in elderly inpatients based on machine learning，to construct an optimized risk prediction model，and to provide a tool for early screening of sarcopenic obesity.

Methods

A retrospective analysis was performed for the data of the elderly inpatients who were treated in The First Affiliated Hospital of Chongqing Medical University from March 2016 to June 2023. The univariate analysis and the least absolute shrinkage and selection operator regression analysis were used to identify related variables，and the multivariate logistic regression analysis was used to investigate the influencing factors for sarcopenic obesity. Four predictive models were constructed，i.e.，logistic regression model，eXtreme Gradient Boosting（XGboost） model，random forest（RF） model，and back propagation neural network（BPNN） model. The optimal model was identified，and Shapley Additive Explanation（SHAP） was used for model interpretation.

Results

The detection rate of sarcopenic obesity was 26.41% among the elderly inpatients. Pre-frailty （odds ratio［

］=2.493，

0.05），frailty（

=4.091，

0.05），age-adjusted Charlson comorbidity index（

=1.110，

0.05），cognitive impairment （

=1.482，

0.05），percentage of body fat（

=1.114，

0.05），and visceral fat area（

=1.014，

0.05） were identified as independent risk factors for sarcopenic obesity in elderly inpatients. The BPNN model showed the best predictive performance，with an area under the receiver operating characteristic curve of 0.842 in the test set，and this model had a Brier score of 0.144，suggesting that it had good performance. The clinical decision curve analysis showed that the model achieved maximum net benefit within the threshold range. SHAP visualization showed that the top three features in terms of

importance were visceral fat area，absence of frailty，and percentage of body fat.

Conclusion

The BPNN predictive model is constructed to provide a methodological reference for identifying the population at a high risk of sarcopenic obesity. It is recommended to focus on the elderly inpatients with cognitive impairment，percentage of body fat ≥35%，visceral fat area ≥100 cm

，or abnormal frailty assessment，and to implement comprehensive interventions for these high-risk individuals

关键词

Keywords

references

刘妍慧，陈树春 . 2022年欧洲临床营养与代谢学会和欧洲肥胖研究学会《肌肉减少性肥胖的定义和诊断标准共识》解读及启示 [J ] . 中国全科医学， 2023 ， 26 （ 12 ）： 1422 - 1428 ．

Liu YH ， Chen SC . Interpretation and clinical significance of the definition and diagnostic criteria for sarcopenic obesity：ESPEN and EASO consensus statement [J ] . Chin Gen Pract ， 2023 ， 26 （ 12 ）： 1422 - 1428 ．

Evans K ， Abdelhafiz D ， Abdelhafiz AH . Sarcopenic obesity as a determinant of cardiovascular disease risk in older people： a systematic review [J ] . Postgrad Med ， 2021 ， 133 （ 8 ）： 831 - 842 ．

Atkins JL ， Whincup PH ， Morris RW ， et al . Sarcopenic obesity and risk of cardiovascular disease and mortality：a population-based cohort study of older men [J ] . J Am Geriatr Soc ， 2014 ， 62 （ 2 ）： 253 - 260 ．

Xu YH ， Song TR ， Cao L ， et al . Impact of sarcopenic obesity on postoperative outcomes in patients with IBD after bowel resection surgery： a retrospective cohort study [J ] . Dis Colon Rectum ， 2026 ， 69 （ 1 ）： 53 - 63 ．

Xia MF ， Chen LY ， Wu L ， et al . Sarcopenia，sarcopenic overweight/obesity and risk of cardiovascular disease and cardiac arrhythmia：a cross-sectional study [J ] . Clin Nutr ， 2021 ， 40 （ 2 ）： 571 - 580 ．

Öztürk ZA ， Türkbeyler İH ， Abiyev A ， et al . Health-related quality of life and fall risk associated with age-related body composition changes；sarcopenia，obesity and sarcopenic obesity [J ] . Intern Med J ， 2018 ， 48 （ 8 ）： 973 - 981 ．

Seo HS ， Lee H ， Kim S ， et al . Paravertebral muscles as indexes of sarcopenia and sarcopenic obesity：comparison with imaging and muscle function indexes and impact on cardiovascular and metabolic disorders [J ] . AJR Am J Roentgenol ， 2021 ， 216 （ 6 ）： 1596 - 1606 ．

Silay K ， Selvi Oztorun H . Sarcopenic obesity is linked to worse clinical outcomes than sarcopenia or obesity alone in hospitalized older adults [J ] . BMC Geriatr ， 2025 ， 25 （ 1 ）： 443 ．

Hirani V ， Naganathan V ， Blyth F ， et al . Longitudinal associations between body composition，sarcopenic obesity and outcomes of frailty，disability，institutionalisation and mortality in community-dwelling older men： The Concord Health and Ageing in Men Project [J ] . Age Ageing ， 2017 ， 46 （ 3 ）： 413 - 420 ．

Luo X ， Cai B ， Jin WW . The prevalence rate of adult sarcopenic obesity and correlation of appendicular skeletal muscle mass index with body mass index，percent body fat，waist-hip ratio，basal metabolic rate，and visceral fat area [J ] . Metab Syndr Relat Disord ， 2023 ， 21 （ 1 ）： 48 - 56 ．

Jiang ML ， Ren X ， Han LY ， et al . Associations between sarcopenic obesity and risk of cardiovascular disease：a population-based cohort study among middle-aged and older adults using the CHARLS [J ] . Clin Nutr ， 2024 ， 43 （ 3 ）： 796 - 802 ．

Charlson M ， Szatrowski TP ， Peterson J ， et al . Validation of a combined comorbidity index [J ] . J Clin Epidemiol ， 1994 ， 47 （ 11 ）： 1245 - 1251 .

Brañez-Condorena A ， Soriano-Moreno DR ， Navarro-Flores A ， et al . Accuracy of the Geriatric Depression Scale（GDS）-4 and GDS-5 for the screening of depression among older adults：a systematic review and meta-analysis [J ] . PLoS One ， 2021 ， 16 （ 7 ）： e0253899 ．

刘　蓓 . 基于综合评估的高龄老人的临床特点分析 [D ] . 重庆：重庆医科大学， 2019 ．

Liu B . Analysis of advanced ages’ clinical characteristics based on comprehensive geriatric assessment [D ] . Chongqing ： Chongqing Medical University ， 2019 ．

赵元萍，丁　睿，谢　红 . 中国版日常生活活动能力量表编制与信效度验证 [J ] . 实用老年医学， 2022 ， 36 （ 12 ）： 1215 - 1219 ．

Zhao YP ， Ding R ， Xie H . Development and validation of reliability and validity of Chinese version of activities of daily living scale [J ] . Pract Geriatr ， 2022 ， 36 （ 12 ）： 1215 - 1219 ．

周文婷，侯韦莲，李　毅 . 微型营养评价对高龄住院病人生存结局的预测价值研究 [J ] . 肠外与肠内营养， 2024 ， 31 （ 6 ）： 321 - 328 ．

Zhou WT ， Hou WL ， Li Y . The predictive value of mini-nutritional assessment on survival outcome of elderly inpatients [J ] . Parenter Enter Nutr ， 2024 ， 31 （ 6 ）： 321 - 328 ．

Jia XF ， Wang ZH ， Huang FF ， et al . A comparison of the Mini-Mental State Examination （MMSE） with the Montreal Cognitive Assessment （ MoCA ） for mild cognitive impairment screening in Chinese middle-aged and older population：a cross-sectional study [J ] . BMC Psychiatry ， 2021 ， 21 （ 1 ）： 485 ．

Li H ， Shen HF ， Yu L ， et al . Cognitive impairment mediates the association between sarcopenic obesity and daily living ability in the frail older population [J ] . Exp Gerontol ， 2025 ， 209 ： 112833 ．

Guo JL ， He QH ， Li YH . Machine learning-based online web calculator predicts the risk of sarcopenic obesity in older adults [J ] . Aging Clin Exp Res ， 2025 ， 37 （ 1 ）： 210 ．

Peng L ， Xiang QW ， Jia G ， et al . Association between sarcopenic obesity and dementia in the Chinese elderly using different definitions of obesity：evidence from the CHARLS [J ] . Front Aging Neurosci ， 2025 ， 17 ： 1540272 ．

Yao GX ， Ma XH ， Wan XY ， et al . Association between sarcopenic obesity and risk of frailty in older adults：a systematic review and meta-analysis [J ] . Age Ageing ， 2025 ， 54 （ 1 ）： afae286 ．

Wei SB ， Nguyen TT ， Zhang Y ， et al . Sarcopenic obesity： epidemiology，pathophysiology，cardiovascular disease，mortality，and management [J ] . Front Endocrinol ， 2023 ， 14 ： 1185221 ．

Feng ZY ， Xia JY ， Yu JH ， et al . Pathophysiological mechanisms underlying sarcopenia and sarcopenic obesity：a systematic review and meta-analysis of biomarker evidence [J ] . Int J Mol Sci ， 2025 ， 26 （ 11 ）： 5113 ．

Tang CB ， Song Y ， Tao FR ， et al . Association of sarcopenic obesity with cognitive dysfunction and dementia：a systematic review and meta-analysis [J ] . Aging Clin Exp Res ， 2025 ， 37 （ 1 ）： 318 ．

Someya Y ， Tamura Y ， Kaga H ， et al . Sarcopenic obesity is associated with cognitive impairment in community-dwelling older adults： The Bunkyo Health Study [J ] . Clin Nutr ， 2022 ， 41 （ 5 ）： 1046 - 1051 ．

Wang H ， Hai S ， Liu YX ， et al . Associations between sarcopenic obesity and cognitive impairment in elderly Chinese community-dwelling individuals [J ] . J Nutr Health Aging ， 2019 ， 23 （ 1 ）： 14 - 20 ．

袁维霞，敖盼盼，马芸，等 . 年龄校正的Charlson合并症指数对老年患者肌少症的预测价值研究 [J ] . 中国全科医学， 2026 ， 29 （ 13 ）： 1673 - 1681 .

Yuan WX ， Ao PP ， Ma Y ， et al . Predictive Value of Age-adjusted Charlson Comorbidity Index for Sarcopenia in Elderly Patients [J ] . Chinese General Practice ， 2026 ， 29 （ 13 ）： 1673 - 1681 .

Deguchi N ， Tanaka R ， Akita T . Association between sarcopenic obesity and frailty risk in community-dwelling older women with locomotive syndrome：a cross-sectional survey [J ] . Cureus ， 2024 ， 16 （ 7 ）： 612 - 620 .

陈小燕，李　梅，黄　莎，等 . 肌少性肥胖与养老机构老年人死亡风险：基于六种肥胖指标的前瞻性队列研究 [J ] . 中华老年医学杂志， 2025 ， 44 （ 7 ）： 943 - 950 ．

Chen XY ， Li M ， Huang S ， et al . Sarcopenic obesity and mortality risk in nursing home residents： a prospective cohort study based on six obesity indicators [J ] . Chin J Geriatr ， 2025 ， 44 （ 7 ）： 943 - 950 ．

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research advances in artificial intelligence for difficult airway prediction

Construction and validation of a machine learning-based model for predicting the risk of intravenous medication

Loneliness and its influencing factors in patients on maintenance hemodialysis comorbid with anemia：an analysis based on machine learning

Construction of a predictive model for the severity of obstructive sleep apnea in patients with type 2 diabetes mellitus based on estimated glucose disposal rate

Value of various machine learning models in differentiating Alzheimer’s disease from moderate-to-severe cerebral small vessel disease

Related Author

Zhou Guilin

Shao Jianlin

Wang Hongmei

Shan Xuefeng

Yang Yang

Xiao Mingzhao

Zhou Zumu

Li Fanghe

Related Institution

Department of Anesthesiology，The First Affiliated Hospital of Kunming Medical University

Department of Urology，The First Affiliated Hospital of Chongqing Medical University

Department of Pharmacy，Bishan Hospital of Chongqing Medical University

Department of Pharmacy，The First Affiliated Hospital of Chongqing Medical University

Department of Anesthesiology，The First Affiliated Hospital of Chongqing Medical University

⁰