The objective of the current study was to provide a method for estimating the spatio-temporal gait parameters using the inertial measurements on the wrist.
Eight younger adults (age = 24.7±2.8 years, weight = 62.3±6.9 kg, height = 171.7±4.7 cm) were recruited in a laboratory study. One IMU (Inertia-Link, MicroStrain, Inc., USA) was placed on the lateral side of the right wrist. The synchronization between IMU and force-plates was achieved by a custom-made Labview program (National Instruments, USA) using time-synchronization triger. All subjects reported right side to be dominant side. The Inertia-Link is a miniature orientation sensor, which is capable of measuring 3D orientation, 3D acceleration, and 3D angular velocity. With respect to the orientation performance, the Inertia-Link has an angular resolution of < 0.1˚, static accuracy of ±0.5˚ and dynamic accuracy of ±2.0˚ RMS (MicroStrain Inc., 2007). The dynamic ranges for the acceleration and angular velocity outputs are ±300˚/s and 5g, respectively (MicroStrain Inc., 2007). The sampling rate was 100Hz.
The IMU on the wrist generated three types of measurements: acceleration, angular velocity (from gyroscope), and orientation. The wrist acceleration presents a distinguishable cyclic pattern (Fig.1) in vertical (VT) direction, where the peak-to-peak duration roughly corresponded to one step (1 step = 0.5 stride). The wrist gyroscope measurements in AP direction show similar repetitive pattern, with peak-to-peak duration roughly corresponding to one stride. With respect to the wrist orientation, peak angles appear to occur simultaneously with the contralateral heel contacts.
For temporal parameters, a strong and significant linear correlation (p < 0.0001, R2 = 0.99) was found between TB and TAP (Fig.2). Based on the temporal parameter prediction model (Eq. 1), the absolute residual error between TB and TAP was, in average, 0.0069s or 0.6% of the mean stride time (1.08s). The TB and TVT, however, had a rather weak and insignificant linear correlation (R2 = 0.33, p < 0.1063).
Current study provided the evidence that inertial signals on the wrist can be used to predict spatio-temporal gait parameters. This finding is the first step towards designing a wrist-based gait-monitoring device with superior user compliance. By allowing an individual’s gait to be assessed outside the laboratory, Such technology will dramatically expand the clinical usefulness of gait analysis, and enable gait analysis to move from being primarily a research and diagnostic tool to a monitoring tool that will facilitate longitudinal tracking of an individual’s gait, which have implications in rehabilitation and remote health care.
Chonbuk National University
Sukwon Kim completed his Ph.D in Industrial and Systems Engineering (Biomedical Engineering option), Virginia Tech. He now teaches sports biomechanics at Chonbuk National University. He is interested in gait mechanisms and slips and falls accident of the elderly. The ultimate goal of his research is to provide information that lead to fall injury reduction or fall prevention.
JeeHoon Sohn has completed Ph.D from KookMin University in Sports Science (biomechanics). He teaches biomechanics at Jeonju University.
Jian Liu completed Ph.D in Industrial and Systems Engineering (specialty in Human Factors Engineering), Virginia Tech. Now, He teaches Safety Engineering at Marshall University.