Classification of Older and Fall-Experienced Subjects by Postural Sway Data Using Mass Spring Damper Model

Area of Science:

• Biomedical Engineering
• Biomechanics
• Gerontology

Background:

• The quiet standing test assesses postural control, but traditional center of pressure (COP) statistics for older adults are insufficient for identifying structural issues.
• Mathematical modeling of COP trajectories with parameters like stiffness and viscosity has potential but requires validation for classifying older and fall-experienced individuals.

Purpose of the Study:

• To evaluate the effectiveness of mass-spring-damper (MSD) model parameters in classifying older and fall-experienced subjects using quiet standing test data.
• To compare the classification accuracy of MSD parameters against conventional descriptive statistics.

Main Methods:

• Six structural parameters of an MSD model were estimated from quiet standing tests of 212 subjects under four conditions.
• A random forest algorithm was employed to classify subjects using estimated MSD parameters and conventional descriptive statistics.

Main Results:

• The MSD parameter method achieved the highest classification accuracy for older subjects in the foam condition (positive likelihood ratio ~8.0).
• For fall-experienced subjects, the MSD parameter method showed a positive likelihood ratio of 5.0, outperforming conventional methods.
• MSD parameters indicated that aging and altered sensory conditions (floor surface, vision) influence COP oscillations.

Conclusions:

• MSD model parameters enhance the classification accuracy of older subjects in quiet standing tests compared to traditional statistics.
• While improvements were noted for fall-experienced subjects, further refinement of the MSD parameter approach is needed for optimal classification.