Gait variability analysis through phase portrait estimated from the Hilbert transform

Abstract

Gait variability has been used to evaluate the ability to control gait. Several studies approached this topic by analysing the influence of different conditions on gait variability, such as different walk speeds, inclined surfaces, load carriage, or comparing characteristics of subject groups, such as age, sedentarism and impairment level. The aim of this study was to develop and assess a new method, based on the property of the Hilbert transform of easily creating a phase portrait from a single time series, capable of estimating variability within gait cycles. The obtained results were based on a comparison of the proposed method with a traditional one whilst analysing a data set related to gait evaluation on inclined surfaces. Furthermore, the influence of noise over the estimated gait variability was assessed. The results showed that the proposed method is less sensitive to the presence of noise, with the advantage of not relying on signal interpolation, being thus an alternative to the analysis of gait variability.