Related Experiment Video
Updated: Oct 17, 2025
13:35
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
Published on: March 21, 2021
10.9K
Predicting Sarcopenia of Female Elderly from Physical Activity Performance Measurement Using Machine Learning
Jeong Bae Ko1, Kwang Bok Kim1, Young Sub Shin1
1Digital Health Care R&D Department, Korea Institute of Industrial Technology, Cheonan, Chuncheongnam-do, South Korea.
Clinical Interventions in Aging
|October 6, 2021
Summary
This study developed a machine learning model to predict sarcopenia in elderly women using physical performance data. The k-nearest neighborhood algorithm achieved 88% accuracy, paving the way for self-monitoring technologies.
Area of Science:
- Gerontology
- Biomedical Engineering
- Machine Learning
Background:
- Sarcopenia, characterized by age-related muscle mass loss, impacts elderly health.
- Early detection is crucial for managing sarcopenia and maintaining quality of life.
Purpose of the Study:
- To develop a machine learning classification model for sarcopenia prediction.
- To utilize inertial measurement unit (IMU)-based physical performance data from elderly females.
Main Methods:
- Seventy-eight elderly females (mean age 78.8 years) participated.
- Timed-Up-and-Go (TUG) and 6-Minute Walk Test (6mWT) were performed with a single IMU.
- 132 features were extracted, selected using the Kruskal-Wallis test, and classified using three ML algorithms.
Main Results:
- The k-nearest neighborhood (kNN) algorithm demonstrated the highest classification performance.
- The kNN model achieved 88% accuracy in predicting sarcopenia.
- The best model utilized 40 key features from TUG and 6mWT data.
Conclusions:
- The developed machine learning model shows promise for sarcopenia prediction in elderly women.
- This research provides a foundation for creating self-monitoring tools for sarcopenia.
- IMU-based physical performance analysis is a viable method for assessing sarcopenia risk.
