A novel approach for assessing gait using foot mounted accelerometers

Abstract

Accelerometer technology is becoming increasingly smaller and cheaper to develop. As a result such devices can easily be integrated into a shoe to ubiquitously capture gait information which could potentially be used to detect development of injuries, neuro-degenerative diseases or a change in disease symptoms. Much research has been done in the past comparing accelerometer data to kinematic or spatio-temporal data; however little has been done investigating what insights into normal and dysfunctional movement patterns accelerometer data from the foot can provide. It is important to first gain an understanding of how foot accelerometer data behaves during healthy gait before developing methods to assess dysfunctional gait with such a tool. In this preliminary study we have analyzed data harnessed from tri-axial accelerometers mounted on the dorsi of the feet in 6 healthy subjects walking at different speeds (slow, normal relaxed walking, fast) to hypothesize what insights into movement and motor control accelerometer data output alone can provide. Results indicate that peak acceleration during initial swing, mean acceleration during mid-swing and peak acceleration at initial contact are main quantitative features that distinguish between walking velocities. These results suggest that quantifying specific acceleration patterns during gait may one day be useful to cheaply and easily detect gait pattern changes due to disease or injury. Though these preliminary results are promising, further work is required to investigate the utility of accelerometer use in a patient population.