Estimation of Temporal Gait Events from a Single Accelerometer Through the Scale-Space Filtering Idea

Area of Science:

• Biomechanics
• Wearable Technology
• Gerontology

Background:

• Gait analysis is crucial for assessing mobility and fall risk, especially in older adults.
• Traditional gait analysis often requires specialized equipment and laboratory settings.
• Developing accessible, wearable systems for gait event demarcation is a significant need.

Purpose of the Study:

• To develop and validate an accelerometry system for precise gait event demarcation and temporal parameter calculation.
• To utilize a single, waist-mounted mobile phone for capturing trunk accelerations during walking.
• To assess the system's capability in differentiating gait characteristics between pre-frail older adults and young healthy individuals.

Main Methods:

• Acquisition of trunk accelerations using a mobile phone placed over the L2 vertebra during walking.
• Application of various filters (e.g., Gaussian filters) to smooth acceleration magnitude and vertical acceleration signals.
• Identification of gait events by detecting peaks in filtered signals, robust against noise and temporal variations.
• Recruitment of five pre-frail older adults and five young healthy adults for experimental validation.

Main Results:

• The system successfully performed gait event demarcation using filtered acceleration data.
• Temporal gait parameters including cadence, step/stride time, and variability were accurately quantified.
• Phase analysis (stance/swing, single/double support) was successfully computed for both participant groups.
• The system demonstrated sensitivity to mobility differences between the pre-frail and young healthy groups.

Conclusions:

• A single, waist-mounted mobile phone can serve as an effective accelerometry system for gait analysis.
• The developed signal processing techniques enable robust gait event identification and temporal parameter calculation.
• This system offers a promising, accessible tool for monitoring gait in diverse populations, including those with reduced mobility.