Related Experiment Video
Updated: Nov 16, 2025
07:52
Evaluating Postural Control and Lower-extremity Muscle Activation in Individuals with Chronic Ankle Instability
Published on: September 18, 2020
8.8K
Classification of high knee flexion postures using EMG signals
Annemarie F Laudanski1, Stacey M Acker1
1Department of Kinesiology, Faculty of Health, University of Waterloo, Waterloo, ON, Canada.
Work (Reading, Mass.)
|February 22, 2021
Summary
Detecting high knee flexion postures using electromyography (EMG) is possible when algorithms are trained on individual data. This approach may help quantify occupational exposures and understand postural demands.
Area of Science:
- Biomechanics
- Occupational Health
- Signal Processing
Background:
- High knee flexion postures in occupational settings increase osteoarthritis risk.
- Electromyographic (EMG) signal pattern recognition can detect and quantify workplace exposures.
- Childcare occupations frequently involve high knee flexion activities.
Purpose of the Study:
- Develop a k-Nearest Neighbor (kNN) algorithm to classify eight common high knee flexion activities in childcare.
- Utilize wireless EMG signals for objective measurement of occupational postures.
Main Methods:
- Recorded EMG signals from eight lower limb muscles of 30 participants.
- Decomposed EMG signals into time- and frequency-domain features.
- Reduced features using neighborhood component analysis to ten time-domain features for kNN classification.
Main Results:
- The kNN algorithm achieved 80.1% accuracy in classifying high knee flexion postures using training data from participants.
- Accuracy dropped to 18.4% when predicting postures from novel subjects not included in the training data.
Conclusions:
- EMG-based classification of high flexion postures is feasible in occupational settings with individualized model training.
- The developed algorithm offers quantitative insights into occupation-specific postural requirements.
- Further research may refine algorithms for broader applicability across different individuals and occupations.
