Related Experiment Video
Updated: Dec 26, 2025
08:56
Clinical Assessment of Spatiotemporal Gait Parameters in Patients and Older Adults
Published on: November 7, 2014
14.2K
The detection of age groups by dynamic gait outcomes using machine learning approaches
Yuhan Zhou1, Robbin Romijnders2, Clint Hansen2
1Center for Human Movement Sciences, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands. y.zhou01@umcg.nl.
Scientific Reports
|March 12, 2020
Summary
Gait analysis using machine learning accurately classifies individuals by age group. Artificial Neural Networks (ANN) show high accuracy in identifying mobility decline and fall risks in older adults.
Area of Science:
- Biomechanics
- Gerontology
- Machine Learning
Background:
- Gait impairments are common in aging, increasing fall risk and reducing independence.
- Geriatric patients face a higher incidence of gait disorders, necessitating clinical assessment.
Purpose of the Study:
- To classify healthy young-middle aged adults, older adults, and geriatric patients using dynamic gait outcomes.
- To compare the classification performance of three supervised machine learning methods.
Main Methods:
- 23 dynamic gait outcomes were derived from trunk 3D-accelerations of 239 subjects.
- Kernel Principal Component Analysis (KPCA) for Support Vector Machine (SVM) dimensionality reduction.
- Random Forest (RF) and Artificial Neural Network (ANN) applied directly to gait outcomes.
Main Results:
- ANN achieved 90% classification accuracy, SVM achieved 89%, and RF achieved 73%.
- Key gait outcomes for classification included Root Mean Square, Cross Entropy, Lyapunov Exponent, step regularity, and gait speed.
- ANN demonstrated automated data reduction and identification of significant gait outcomes.
Conclusions:
- ANN is a preferred method for gait classification due to its efficiency and accuracy.
- Identified gait outcomes can aid clinicians in diagnosing mobility issues, fall risk, and monitoring interventions.
