addressing sarcopenia-related public health burden in institutionalised older adults through scalable gait-based care surveillance: a machine learning approach
1
School of Public Health, Shantou University, Shantou, China
2
Department of Orthopedics, Guangdong Provincial People's Hospital (Guangdong Academy of Medical Sciences), Guangzhou, China
3
Taikang Xianlin Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China
Popul. Med. 2026;8(Supplement Supplement 1):
ABSTRACT
INTRODUCTION:
sarcopenia is a prevalent geriatric syndrome associated with falls disability and excess health care utilisation among institutionalised older adults.¹,² existing screening and diagnostic approaches are constrained by operator dependence limited accessibility and poor scalability at the population level.³ gait characteristics may provide an objective non-invasive alternative for sarcopenia surveillance.⁴,⁵ this study aimed to evaluate whether phase-specific six-degree-of-freedom lower-limb gait kinematics combined with machine learning can support sarcopenia detection in nursing home populations.
METHODS:
this cross-sectional study included 299 adults aged 65 years or older from two nursing homes in china. sarcopenia was defined according to the asian working group for sarcopenia 2019 criteria.⁶ bilateral six-degree-of-freedom lower-limb kinematics were collected during treadmill walking at 2 km per hour. candidate features were selected using univariate screening and lasso regression. data were split into training and validation sets at a ratio of 7 to 3. nine machine learning models were developed using ten-fold cross-validation. model performance was assessed using area under the receiver operating characteristic curve area under the precision recall curve calibration and decision curve analysis.
RESULTS:
the prevalence of sarcopenia was 38.5 percent. individuals with sarcopenia had lower body mass index and phase angle compared with non-sarcopenic participants.⁶,⁷ a simplified lightgbm model incorporating eight predictors achieved the best performance with an area under the receiver operating characteristic curve of 0.888 and an area under the precision recall curve of 0.831. selected gait kinematic features showed significant correlations with conventional physical performance and body composition indicators.⁴,⁸ a prototype application was developed for individual risk estimation.
CONCLUSIONS:
phase-specific six-degree-of-freedom gait kinematics combined with interpretable machine learning offer a non-invasive and scalable approach for sarcopenia screening and monitoring in institutionalised older adults. this approach has value for population-based surveillance and early intervention strategies in aging societies.²,⁵
sarcopenia is a prevalent geriatric syndrome associated with falls disability and excess health care utilisation among institutionalised older adults.¹,² existing screening and diagnostic approaches are constrained by operator dependence limited accessibility and poor scalability at the population level.³ gait characteristics may provide an objective non-invasive alternative for sarcopenia surveillance.⁴,⁵ this study aimed to evaluate whether phase-specific six-degree-of-freedom lower-limb gait kinematics combined with machine learning can support sarcopenia detection in nursing home populations.
METHODS:
this cross-sectional study included 299 adults aged 65 years or older from two nursing homes in china. sarcopenia was defined according to the asian working group for sarcopenia 2019 criteria.⁶ bilateral six-degree-of-freedom lower-limb kinematics were collected during treadmill walking at 2 km per hour. candidate features were selected using univariate screening and lasso regression. data were split into training and validation sets at a ratio of 7 to 3. nine machine learning models were developed using ten-fold cross-validation. model performance was assessed using area under the receiver operating characteristic curve area under the precision recall curve calibration and decision curve analysis.
RESULTS:
the prevalence of sarcopenia was 38.5 percent. individuals with sarcopenia had lower body mass index and phase angle compared with non-sarcopenic participants.⁶,⁷ a simplified lightgbm model incorporating eight predictors achieved the best performance with an area under the receiver operating characteristic curve of 0.888 and an area under the precision recall curve of 0.831. selected gait kinematic features showed significant correlations with conventional physical performance and body composition indicators.⁴,⁸ a prototype application was developed for individual risk estimation.
CONCLUSIONS:
phase-specific six-degree-of-freedom gait kinematics combined with interpretable machine learning offer a non-invasive and scalable approach for sarcopenia screening and monitoring in institutionalised older adults. this approach has value for population-based surveillance and early intervention strategies in aging societies.²,⁵
| eISSN: | 2654-1459 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
