Related Experiment Videos
An autoregressive modeling approach to analyzing wheelchair propulsion forces
A M Koontz1, R A Cooper, M L Boninger
1Department of Rehabilitation Science and Technology, University of Pittsburgh, 5044 Forbes Tower, Pennsylvania 15261, Pittsburgh, USA. amkst63@pitt.edu
Medical Engineering & Physics
|June 28, 2001
Summary
Autoregressive (AR) modeling offers a more sensitive method for analyzing wheelchair propulsion, detecting subtle asymmetries in force waveforms that traditional analyses miss. This advanced technique helps identify potential causes of joint pain and injury in wheelchair users.
Area of Science:
- Biomechanics
- Rehabilitation Engineering
Background:
- Traditional analysis of wheelchair propulsion relies on averages and peak values, which may oversimplify complex biomechanical signals.
- This approach may not effectively identify factors contributing to joint pain or injury in wheelchair users.
Purpose of the Study:
- To introduce and evaluate autoregressive (AR) modeling as a novel system identification technique for analyzing wheelchair propulsion force waveforms.
- To compare the sensitivity of AR modeling against point-wise methods in detecting asymmetries in propulsion technique.
Main Methods:
- Autoregressive (AR) modeling was employed to create a model force waveform from digital pushrim force data.
- Twenty wheelchair users propelled at a constant velocity (0.9 m/s) on a roller system, with 20-second force data collected using SMART(Wheel)s.
Main Results:
- While group analysis indicated even propulsion, individual AR model error estimates revealed significant waveform asymmetry in 25% of participants.
- Point-wise methods (means and variances) failed to detect these individual asymmetries, potentially misclassifying propulsion technique.
Conclusions:
- AR modeling is a more sensitive approach for detecting anomalies and asymmetries in individual wheelchair propulsion technique compared to traditional point-wise methods.
- This advanced analysis can provide deeper insights into biomechanical patterns that may predispose users to joint pain or injury.