Quantitative falls risk estimation through multi-sensor assessment of standing balance

Area of Science:

• Gerontology
• Biomedical Engineering
• Rehabilitation Science

Background:

• Falls are a major cause of injury, hospitalization, and death in older adults globally.
• Postural stability measures are linked to fall incidence in this population.
• Current falls risk assessment methods have limitations in accessibility and accuracy.

Purpose of the Study:

• To develop and validate a novel multi-sensor quantitative balance metric model for classifying fallers and non-fallers.
• To compare the model's classification accuracy against the established Berg balance scale.
• To create a deployable system for falls risk assessment in home or clinical settings.

Main Methods:

• Data collected from 120 community-dwelling older adults (65 fallers, 55 non-fallers) using pressure-sensitive platforms and body-worn inertial sensors.
• Participants performed standing balance tasks in a geriatric research clinic.
• A support vector machine model was employed to classify fall history.

Main Results:

• The multi-sensor model achieved a mean classification accuracy of 73.07% for identifying participants with a history of falls.
• Classification accuracy was slightly higher when analyzing male (72.80%) and female (73.33%) data separately.
• The developed model significantly outperformed the Berg balance scale (59.42% accuracy).

Conclusions:

• Novel multi-sensor balance metrics can accurately classify falls risk in older adults.
• This approach offers a robust and potentially more accurate alternative to traditional falls risk assessment tools.
• The findings support the development of easily deployable systems for unsupervised falls risk monitoring.