Improving the accuracy of pedometer used by the elderly with the FFT algorithm.

Area of Science:

• Gerontology
• Biomedical Engineering
• Kinesiology

Background:

• Accelerometer-type pedometers often show inaccuracies in older adults, particularly those with slow walking speeds or gait disorders.
• Accurate step counting is crucial for monitoring physical activity and health in the elderly population.

Purpose of the Study:

• To investigate and enhance the accuracy of accelerometer-based pedometers for elderly individuals.
• To develop a reliable method for estimating step counts in older adults with varying walking speeds and gait characteristics, including cane users.

Main Methods:

• Employed a novel approach using triaxial accelerometer signals and fast Fourier transform (FFT) to determine cadence and estimate step counts.
• Compared pedometer (PM) and FFT-derived (FM) step counts against visually counted actual steps (RM) in elderly participants (N=49) walking with or without a cane.

Main Results:

• Standard pedometer measurements (PM) significantly underestimated actual steps (RM) with a large error (53.2 +/- 34.1%).
• The proposed FFT-based method (FM) demonstrated high accuracy, with no significant difference from actual steps (RM) and a low average error (-0.7 +/- 7.9%).

Conclusions:

• The fast Fourier transform (FFT) method is a suitable and accurate technique for estimating the number of steps during walking in the elderly population.
• This improved method holds potential for more reliable physical activity monitoring in older adults, including those with gait impairments.